JP2018191761A - Information processing apparatus, information processing method, and program - Google Patents

Abstract

An information processing apparatus, method, and program for detecting the presence of artifacts due to shadows in a motion contrast image are provided. An acquisition means for acquiring a plurality of tomographic data obtained based on measurement light controlled to scan the same position of a subject, and motion contrast data of the subject is calculated from the plurality of tomographic data. A calculation means; a mask means for masking motion contrast data based on at least one tomographic data of the plurality of tomographic data; a generation means for generating a motion contrast image based on the motion contrast data after the mask processing; Based on at least one tomographic data of the plurality of tomographic data, a specifying means for specifying a shadow portion of the subject, a specifying result of the specifying means, and a position where the motion contrast data is masked by the masking means, a motion contrast image Means for detecting artifacts in Equipped with a. [Selection] Figure 9

Description

  The present invention relates to an information processing apparatus, an information processing method, and a program.

  An optical coherence tomography (OCT) using optical coherence tomography (OCT) has been put into practical use as a method for non-destructively and non-invasively acquiring a tomographic image of a measurement target such as a living body. . The OCT apparatus is widely used especially in ophthalmic diagnosis.

  In OCT, a tomographic image of a measurement object can be obtained by causing the light reflected from the measurement object to interfere with the light reflected from the reference mirror and analyzing the intensity of the interference light. As such OCT, time domain OCT (TD-OCT: Time Domain OCT) and Fourier domain OCT (FD-OCT: Fourier Domain OCT) are known. In the time domain OCT, the depth information of the measurement object is obtained by changing the position of the reference mirror. In the Fourier domain OCT, the depth information of the subject is replaced with frequency information and acquired. The Fourier domain OCT includes a spectral domain OCT (SD-OCT: Spectral Domain OCT) and a wavelength sweep type OCT (SS-OCT: Swept Source OCT). In the spectral domain OCT, interference light is dispersed and depth information is replaced with frequency information. On the other hand, in the wavelength sweep type OCT, the wavelength is first dispersed and output, and the depth information is obtained by replacing it with the frequency information.

  In recent years, angiographic methods using these OCTs without using contrast agents have been proposed. This angiography method is called OCT angiography (OCTA), and the obtained OCTA image is a kind of motion contrast image obtained by imaging motion contrast data. Here, the motion contrast data is data obtained by repeatedly photographing the same cross section of the subject and detecting a temporal change of the subject during the photographing. The motion contrast data is obtained, for example, by calculating the phase difference or vector difference of the complex OCT signal or temporal change in intensity (see Patent Document 1).

  Further, there is a threshold process as a method for reducing noise in the OCTA image. In this threshold processing, the intensity of the complex OCT signal is compared with the threshold, and when the intensity of the complex OCT signal falls below the threshold, the motion contrast data corresponding to the signal is set to 0 (masked). Patent Document 2 discloses such threshold processing. In Patent Document 2, a value 10 dB higher than the average value of the noise floor is used as the threshold value.

  On the other hand, when focusing on the subject's eye as the subject of the OCT apparatus, the subject's eye has a shield (such as retinal blood vessels, vitreous turbidity, retinal hemorrhage, or hard vitiligo) that blocks the measurement light from entering the retina. There is. When these shielding objects exist, the measurement light is blocked by the shielding object, and the measurement light is difficult to reach backward from the shielding object. In this case, in the obtained OCT image, a shadow is generated behind the shielding object in the incident direction of the measurement light, and the image luminance of the OCT image is lowered in a portion where the shadow is generated (shadow portion). For example, since the retinal blood vessel exists from the ganglion cell layer to the vicinity of the boundary between the inner granule layer and the outer reticulated layer, when the retinal blood vessel becomes a shielding object, in the OCT image, the outer portion that is behind the boundary portion. Shading occurs from the reticular layer to the retinal pigment epithelium and choroid. In addition, when vitreous opacity becomes a shield, shadows occur in all layers from the inner boundary membrane to the retinal pigment epithelium and choroid in the OCT image.

  When the OCTA image is generated by performing the noise reduction processing described in Patent Document 2 on the shadow portion where the image luminance of the OCT image is low, the motion contrast data in many portions of the shadow portion is set to 0 (mask ) For this reason, in the obtained OCTA image, a part where the shielding object is not present is imaged, but a part where the shielding object is present is not imaged and is displayed as an artifact.

JP, 2015-131107, A US Patent Application Publication No. 2014/221825

  Focusing on the area where the image brightness of the OCTA image is low, when there is no temporal change of the subject due to the influence of disease or the like and the value of the motion contrast data is low, the motion contrast data is masked by the shadow and displayed as an artifact. There is a case. Since it is difficult to discriminate both in the region where the image brightness of the OCTA image is low, there is a possibility that the diagnosis accuracy of the doctor is lowered.

  Therefore, the present invention provides an information processing apparatus, information processing method, and program capable of detecting the presence of artifacts due to shadows in a motion contrast image.

  An information processing apparatus according to an embodiment of the present invention includes an acquisition unit that acquires a plurality of tomographic data obtained based on measurement light controlled to scan the same position of a subject, and the plurality of tomographic data. Calculation means for calculating motion contrast data of the subject; mask means for masking the motion contrast data based on at least one tomographic data of the plurality of tomographic data; and motion contrast data after the mask processing. A generating unit that generates a motion contrast image, a specifying unit that specifies a shadow portion of the subject based on at least one tomographic data of the plurality of tomographic data, a specifying result of the specifying unit, and the mask Based on the position where the motion contrast data is masked by means And detection means for detecting an artifact in the motion contrast image.

  An information processing method according to another embodiment of the present invention includes a step of acquiring a plurality of tomographic data obtained based on measurement light controlled to scan the same position of a subject, and the plurality of tomographic data. A step of calculating motion contrast data; a step of masking the motion contrast data based on at least one tomographic data of the plurality of tomographic data; and a motion contrast image based on the motion contrast data after the mask processing A step of specifying a shadow portion of the subject based on at least one tomographic data of the plurality of tomographic data, a specifying result of the step of specifying the shadow portion, and a step of performing the mask processing Based on the position where the motion contrast data is masked, the motion And detecting artifacts in emission contrast image.

  According to the present invention, it is possible to detect the presence of an artifact due to a shadow in a motion contrast image.

1 shows a schematic configuration of an OCT apparatus according to Embodiment 1 of the present invention. 1 shows a schematic configuration of an imaging optical system. The schematic structure of a signal processing part and a display part is shown. An example of an anterior segment image and an SLO image displayed on the display unit during observation is shown. An example of a B-scan image of the retina at a certain position in the Y direction is shown. 3 is a flowchart of a series of processes performed by a signal processing unit according to the first embodiment. 6 is a flowchart of motion contrast data generation processing according to the first embodiment. 6 is a flowchart of motion contrast data threshold processing according to the first embodiment. 6 is a flowchart of artifact detection and detection result display processing according to the first embodiment. 6 is a flowchart of artifact position detection processing according to the first embodiment. An example of a display of an OCTA image and the detection result of an artifact is shown. 10 is a flowchart of artifact detection and detection result display processing according to the second embodiment.

  Hereinafter, exemplary embodiments for carrying out the present invention will be described in detail with reference to the drawings. However, dimensions, materials, shapes, and relative positions of components described in the following embodiments are arbitrary and can be changed according to the configuration of the apparatus to which the present invention is applied or various conditions. Also, in the drawings, the same reference numerals are used between the drawings to indicate the same or functionally similar elements.

[Example 1]
Hereinafter, an OCT apparatus according to a first embodiment of the present invention will be described with reference to FIGS. FIG. 1 shows a schematic configuration of an OCT apparatus 50 according to this embodiment. In the following description, it is assumed that a human eye (examined eye) is handled as a subject.

  The OCT apparatus 50 is provided with an imaging optical system 100, a signal processing unit 300 (information processing device), and a display unit 400 (display means). The imaging optical system 100 obtains tomographic information by scanning the measurement light on the eye to be examined and detecting interference light caused by the return light of the measurement light and the reference light. The signal processing unit 300 generates a tomographic image, an OCTA image, and the like of the eye to be examined from the tomographic information acquired by the imaging optical system 100. The display unit 400 displays the tomographic image and OCTA image of the eye to be examined generated by the signal processing unit 300, the subject information received from the signal processing unit 300, and the like.

  Here, the signal processing unit 300 may be configured by using a general-purpose computer, or may be configured by a computer dedicated to the OCT apparatus 50. The display unit 400 may be configured using an arbitrary monitor. In the present embodiment, the imaging optical system 100, the signal processing unit 300, and the display unit 400 are separately configured, but a part or all of them may be integrally configured.

  Next, the configuration of the imaging optical system 100 will be described with reference to FIG. FIG. 2 shows a schematic configuration of the imaging optical system 100.

  In the imaging optical system 100, the objective lens 1 is disposed facing the eye E, and the first dichroic mirror 2 and the second dichroic mirror 3 are disposed on the optical axis thereof. By the first and second dichroic mirrors 2 and 3, the optical path from the objective lens 1 is the optical path L1 of the OCT optical system, the SLO optical system for observing the eye E, the optical path L2 for the fixation lamp, and the front Branches into the optical path L3 for eye part observation for each wavelength band.

  The SLO optical system and the optical path L2 for the fixation lamp are provided with SLO scanning means 4, lenses 5, 6, mirror 7, third dichroic mirror 8, photodiode 9, SLO light source 10, and fixation lamp 11. Yes.

  The SLO scanning means 4 is a scanning means for scanning the light emitted from the SLO light source 10 and the fixation lamp 11 on the eye E, and scans the light in the X direction and the X scanner. Y scanner. In this embodiment, the X scanner is constituted by a polygon mirror that performs high-speed scanning, and the Y scanner is constituted by a galvanometer mirror. The scanner constituting the scanning unit is not limited to this, and any deflecting unit can be used according to a desired configuration.

  The lens 5 can be driven in the optical axis direction indicated by an arrow in the drawing by a motor (not shown) controlled by the signal processing unit 300 for focusing the SLO optical system and the fixation lamp. The mirror 7 is a prism on which a perforated mirror or a hollow mirror is deposited, and separates illumination light from the SLO light source 10 and return light of illumination light from the eye E to be examined. The third dichroic mirror 8 branches the optical path L2 into an optical path to the SLO light source 10 and an optical path to the fixation lamp 11 for each wavelength band.

  The photodiode 9 detects the return light of the illumination light from the eye E. The SLO light source 10 generates light having a wavelength near 780 nm. The fixation lamp 11 emits visible light and is used to promote fixation of the subject.

  The light emitted from the SLO light source 10 is reflected by the third dichroic mirror 8, passes through the mirror 7, passes through the lenses 6 and 5, and is scanned on the eye E by the SLO scanning unit 4. The return light of the illumination light from the eye E is reflected by the mirror 7 after being returned through the same path as the illumination light, and is guided to the photodiode 9. The photodiode 9 detects the return light of the illumination light from the eye E and sends an output signal based on the detected return light to the signal processing unit 300. The signal processing unit 300 can generate a fundus image of the eye E based on the received signal.

  The fixation lamp 11 passes through the third dichroic mirror 8 and the mirror 7, passes through the lenses 6 and 5, and is scanned on the eye E by the SLO scanning unit 4. At this time, the signal processing unit 300 controls the fixation lamp 11 to blink in accordance with the movement of the SLO scanning unit 4, thereby creating an arbitrary shape at an arbitrary position on the eye E to be examined. Can be fixed.

  A lens 12, a split prism 13, a lens 14, and a CCD 15 for observing the anterior segment are disposed in the optical path L3 for observing the anterior segment. The split prism 13 is disposed at a position conjugate with the pupil of the eye E, and the signal processing unit 300 determines the distance between the eye E and the imaging optical system 100 in the Z direction (front-rear direction) and splits the anterior eye. It can be detected using an image.

  The CCD 15 has sensitivity at a wavelength of a light source for anterior ocular segment observation (not shown), specifically around 970 nm. The CCD 15 detects return light emitted from the anterior eye portion observation light source, reflected and scattered by the eye E, and sends an output signal to the signal processing unit 300 based on the detected return light. The signal processing unit 300 can generate an anterior segment image of the eye E based on the received output signal from the CCD 15.

  Next, the optical path L1 of the OCT optical system will be described. An XY scanner 16 and lenses 17 and 18 are arranged in the optical path L1 of the OCT optical system for capturing a tomographic image of the eye E.

  The XY scanner 16 is an OCT scanning unit for scanning light from the OCT light source 20 on the eye E. In this embodiment, the XY scanner 16 is illustrated as a single mirror, but is composed of a galvanometer mirror that performs scanning in the XY biaxial directions. The configuration of the XY scanner 16 is not limited to this, and any deflecting means can be used according to a desired configuration.

  The lens 17 is driven in the optical axis direction indicated by an arrow in the figure by a motor (not shown) controlled by the signal processing unit 300 in order to focus the measurement light emitted from the optical fiber 21 on the eye E. be able to. By this focusing, the return light of the measurement light from the eye E is simultaneously incident on the tip of the optical fiber 21 in the form of a spot. The optical fiber 21, the optical members disposed in the optical path L1, the first and second dichroic mirrors 2 and 3, the objective lens 1 and the like constitute an OCT measurement optical system through which measurement light propagates in the OCT optical system.

  In front of the lens 18, an optical coupler 19, an OCT light source 20, optical fibers 21 to 24 connected to and integrated with the optical coupler 19, a lens 25, a dispersion compensation glass 26, a reference mirror 27, and a spectrometer 28 are arranged. Has been.

  The OCT light source 20 is a light source for generating light (low coherence light). In the present embodiment, an SLD (Super Luminescent Diode) light source is used as the OCT light source 20. The OCT light source 20 is not limited to the SLD, and an arbitrary light source may be used according to a desired configuration.

  The light emitted from the OCT light source 20 is guided to the optical coupler 19 through the optical fiber 22. The optical coupler 19 splits the light from the OCT light source 20 into measurement light and reference light. The measurement light is emitted from the optical fiber 21 toward the eye E through the objective lens 1 on the optical path L1 of the OCT optical system. The measurement light emitted toward the eye E is reflected and scattered by the eye E and reaches the optical coupler 19 through the same optical path.

  On the other hand, the reference light is emitted from the optical fiber 23 toward the reference mirror 27 through the lens 25 and the dispersion compensation glass 26. The reference light reflected from the reference mirror 27 reaches the optical coupler 19 through the same optical path. The dispersion compensation glass 26 can adjust the dispersion of the reference light to match the dispersion of the measurement light and the reference light. Here, the optical fiber 23, the lens 25, the dispersion compensation glass 26, and the reference mirror 27 constitute an OCT reference optical system.

  The measurement light and the reference light that have reached the optical coupler 19 in this way are combined in the optical coupler 19 and become interference light. Here, the measurement light and the reference light cause interference when the optical path length of the measurement light and the optical path length of the reference light are substantially the same. The reference mirror 27 is held by an unillustrated motor and drive mechanism controlled by the signal processing unit 300 so as to be adjustable in the optical axis direction indicated by an arrow in the figure, and has an optical path length of measurement light that varies depending on the eye E to be examined. It is possible to match the optical path length of the reference light. The interference light is guided to the spectroscope 28 via the optical fiber 24.

  The spectroscope 28 is provided with lenses 29 and 31, a diffraction grating 30, and a line sensor 32. The interference light emitted from the optical fiber 24 becomes parallel light through the lens 29, and then is split by the diffraction grating 30 and imaged on the line sensor 32 by the lens 31. The line sensor 32 sends an interference signal to the signal processing unit 300 based on the detected interference light. The signal processing unit 300 can generate a tomographic image and an OCTA image of the eye E based on the received interference signal from the line sensor 32.

  Next, the configuration of the signal processing unit 300 and the display unit 400 will be described with reference to FIG. FIG. 3 shows a schematic configuration of the signal processing unit 300 and the display unit 400. The signal processing unit 300 includes an acquisition unit 310 (acquisition unit), a drive control unit 320, an image generation unit 330, a shadow identification unit 340 (identification unit), an artifact detection unit 350 (detection unit), a storage unit 360, and display control. A unit 370 (display control means) is provided.

  The acquisition unit 310 acquires various signals from the photodiode 9, the CCD 15, and the line sensor 32 of the imaging optical system 100. The acquisition unit 310 can also acquire, from the image generation unit 330, a signal after Fourier transform generated based on the interference signal, a signal obtained by performing some signal processing on the signal, and the like. The drive control unit 320 controls driving of each component of the imaging optical system 100 such as the OCT light source 20 and the XY scanner 16.

  The image generation unit 330 obtains the eye to be examined from a plurality of data based on the output signal from the photodiode 9 when the eye E is scanned in the X direction and the Y direction using the SLO scanning unit 4 acquired by the acquisition unit 310. An SLO image that is a fundus image of E is generated. Further, the image generation unit 330 generates an anterior ocular segment image of the eye E based on the output signal from the CCD 15 acquired by the acquisition unit 310.

  Furthermore, the image generation unit 330 generates a tomographic image based on the interference signal from the line sensor 32 acquired by the acquisition unit 310. More specifically, the image generation unit 330 performs Fourier transform on the interference signal obtained from the line sensor 32 by the series of processes described above, and converts the signal after Fourier transform into luminance or density information. Thereby, the image generation unit 330 acquires a tomographic image in the depth direction (Z direction) at one point of the fundus of the eye E to be examined. Such a scanning method is called A scan, and the obtained tomographic image is called A scan image.

  A plurality of A-scan images can be acquired by repeatedly performing such A-scan while scanning the measurement light in a predetermined transverse direction of the fundus by the XY scanner 16. For example, if the measurement light is scanned in the X direction, a tomographic image on the XZ plane is obtained, and if the measuring light is scanned in the Y direction, a tomographic image on the YZ plane is obtained. A method of scanning the fundus of the eye E in this way in a predetermined transverse direction is called a B scan, and the obtained tomographic image is called a B scan image. Further, a method of scanning the XZ plane or YZ plane of the B scan in a direction perpendicular to the XZ plane is called a C scan, and the obtained three-dimensional tomographic image of XYZ is called a C scan image.

  The image generation unit 330 is provided with an OCTA image generation unit 331 (generation unit). The OCTA image generation unit 331 calculates motion contrast data using tomographic data based on the interference signal, and generates an OCTA image. Here, the tomographic data includes an interference signal of the eye E to be examined, a signal after the Fourier transform of the interference signal, a signal obtained by subjecting the signal after the Fourier transform to arbitrary signal processing for image generation, and luminance based on these signals Includes signals such as values. Note that the image generation unit 330 generates one B-scan image from a set of tomographic data acquired in one B-scan.

  The OCTA image generation unit 331 includes a calculation unit 332 (calculation unit) and a mask processing unit 333 (mask unit). The calculation unit 332 generates motion contrast data based on the tomographic data. More specifically, the calculation unit 332 generates motion contrast data from m sets of tomographic data acquired by m B scans corresponding to m B scan images. The calculation unit 332 can also specify the boundary position of each layer in the tomographic data by performing arbitrary segmentation processing on the tomographic data. The mask processing unit 333 compares the value of the tomographic data with the threshold value, and when the value of the tomographic data is equal to or smaller than the threshold value, masks the motion contrast data at the pixel position corresponding to the tomographic data.

  The shadow specifying unit 340 specifies a portion (shadow part) where a shadow is generated in the tomographic image based on the tomographic data or the motion contrast data. The artifact detection unit 350 detects an artifact caused by the occurrence of a shadow based on the position of the pixel whose motion contrast data is masked by the mask processing unit 333 and the identification result by the shadow identification unit 340.

  The storage unit 360 forms various data such as SLO images, tomographic data, OCTA images, and detected artifact positions generated by the image generation unit 330, information on the input subject, and the signal processing unit 300. Stores programs and the like. The display control unit 370 controls the display unit 400 and causes the display unit 400 to display various images stored in the storage unit 360, information on the subject, and the like. The display control unit 370 can also highlight and display the artifact region detected by the artifact detection unit 350 on the display unit 400.

  Each component of the signal processing unit 300 can be configured by a module executed by the CPU or MPU of the signal processing unit 300. In addition, each component of the signal processing unit 300 may be configured by a circuit that realizes a specific function such as an ASIC. The storage unit 360 can be configured using an arbitrary storage device / storage medium such as a memory or an optical disk.

  Next, from the observation to the imaging of the eye E in the OCT apparatus 50 including the imaging optical system 100, the signal processing unit 300, and the display unit 400 will be described. First, observation will be described with reference to FIG. FIG. 4 shows an anterior segment image 410 and an SLO image 420 that are displayed on the display unit 400 when the eye E is observed.

  First, after the eye E is positioned in front of the objective lens 1, the examiner views the anterior eye image 410 using input means (not shown) such as a joystick for moving the imaging optical system 100. While aligning the eye E and the imaging optical system 100 in the XYZ directions. The examiner aligns the imaging optical system 100 in the X and Y directions so that the pupil center of the anterior segment image 410 is positioned at the center of the display screen of the anterior segment image 410. In addition, when the alignment of the imaging optical system 100 in the Z direction is not appropriate, the anterior segment image 410 is split along the broken line 411, so that the examiner does not split the anterior segment image 410 in the Z direction. Perform position alignment.

  When the alignment of the eye E and the imaging optical system 100 in the XYZ directions is completed, the SLO image 420 generated based on the scanning of the illumination light in the XY directions by the SLO scanning unit 4 is displayed. The anterior eye portion image 410 and the SLO image 420 are updated as needed, and the examiner can observe the eye E without delay.

  A scan area 421 in the SLO image 420 indicates an area to be scanned when acquiring tomographic data used for generating an OCTA image, and is superimposed on the SLO image 420. The examiner operates a scanning position changing means (not shown) such as a mouse or a touch panel in the scanning area 421 to set a desired scanning position. Observation is completed by these operations.

  In this embodiment, the examiner performs the alignment of the eye E and the imaging optical system 100 in the XYZ directions and sets the scan area. However, for example, the signal processing unit 300 may be configured to automatically perform these processes based on the anterior segment image 410, the SLO image 420, information on the imaging region, and the like.

  Next, imaging of the eye E (OCTA imaging) will be described. When the examiner operates an imaging start button (not shown), the imaging optical system 100 scans the scan area 421 with measurement light using the XY scanner 16. In OCTA imaging, B scans of the same cross section, that is, the same position are repeatedly performed, and a temporal change of the subject between the scans is detected. For this reason, in this embodiment, the B scan at the same position of the eye E is repeated m times. Note that the number of repetitions m of the B scan may be set to an arbitrary number of 2 or more according to a desired configuration. In this embodiment, when scanning the scan area 421, the B scan is repeated m times at each of the Y-direction scan positions Y1 to Yn.

  In this embodiment, each B scan is composed of A scan data at p sample positions in the X direction, and one A scan is composed of q data in the Z direction. Therefore, by scanning the scan area 421, n × m B scan data on the XYZ plane (the number of A scan data is n × m × p, and the total number of data is n × m × p × q). Is obtained. Note that n, p, and q may be set to any number according to a desired configuration.

  When the acquisition of data in the scan area 421 is completed by the imaging optical system 100, the image generation unit 330 is a B-scan image that is a tomographic image of the eye E based on the data acquired from the imaging optical system 100 by the acquisition unit 310. Is generated. Specifically, the image generation unit 330 performs background subtraction and Fourier transform for noise reduction on the interference signal obtained in each B scan. Since a complex OCT signal including a real part and an imaginary part is obtained by Fourier transform, the image generation unit 330 can generate a B-scan image by calculating the absolute value of the complex OCT signal. Note that the tomographic image generation method is not limited to this, and the tomographic image may be generated based on the interference signal by any known method.

  When the image generation unit 330 generates a B scan image, the storage unit 360 stores the generated B scan image, and the OCTA image generation unit 331 generates an OCTA image using the generated B scan image. Specific generation processing of the OCTA image will be described later. When the OCTA image generation unit 331 generates the OCTA image, the storage unit 360 stores the generated OCTA image, and the display control unit 370 causes the display unit 400 to display the OCTA image.

  Further, the artifact detection unit 350 detects an artifact caused by a shadow in the OCTA image. Thereafter, the display control unit 370 causes the display unit 400 to highlight and display artifacts due to shadows based on the detection result by the artifact detection unit 350.

  Next, with reference to FIG. 5, a description will be given of shadows caused by a shield that exists in the eye E and that blocks measurement light from the OCT light source 20 from entering the retina. FIG. 5 shows an example of a B-scan image of the retina at a certain position in the Y direction in the scan area 421. In FIG. 5, the upper part of the figure is a vitreous body, and the measurement light from the OCT light source 20 is incident from the upper part of the figure. In the figure, the incident direction 500 of the measurement light is indicated by an arrow.

  In FIG. 5, X70 indicates the position of the tomographic image in the X direction when there is vitreous turbidity (not shown) in the incident direction of the measurement light. X150 indicates the position of the tomographic image in the X direction when there is a retinal blood vessel in the measurement light incident direction. Further, X170 indicates the position in the X direction of the tomographic image when there is no vitreous opacification and retinal blood vessels in the incident direction of the measurement light. The shaded portion is shown with low luminance on the tomographic image due to the lower intensity of the interference signal obtained from the region.

  In the tomographic image at the X-direction position X70, it can be seen that a shadow is generated in the tomography E1 from the inner boundary membrane to the choroid. This is because there is a vitreous turbidity (not shown) in the incident direction of the measurement light, and the measurement light is blocked by the vitreous turbidity. In this case, it is difficult for the measurement light to reach the fault E1 from the inner boundary membrane to the choroid existing behind the vitreous turbidity in the incident direction of the measurement light, and the shadow is reflected on the fault E1 at the rearward position of the vitreous turbidity. Will occur.

  In the tomographic image at the X-direction position X150, the retinal blood vessel E2 can be observed. The retinal blood vessel E2 is generally observed as an elongated high reflector, and exists from the ganglion cell layer to the vicinity of the boundary between the inner granular layer and the outer reticulated layer. When the retinal blood vessel E2 exists, the measurement light is blocked by the retinal blood vessel E2, and as a result, the measurement light does not easily reach the back of the retinal blood vessel E2, and a shadow is generated in the tomography E3 behind the retinal blood vessel E2.

  In the tomographic image at the X-direction position X170, the tomography E4 from the inner boundary membrane to the choroid can be observed without shadow. This is because there is no shield at the X direction position X170 that blocks the measurement light from entering the retina in the measurement light incident direction.

  As described above, when there is a shielding object in the eye E, the shadow is shadowed in the depth direction (Z direction) of the eye E as viewed from the incident direction 500 of the measurement light behind the shielding object. Occurs. Therefore, the position in the Z direction where the shadow is generated varies depending on the position in the Z direction of the eye E to be examined.

  In this embodiment, the X-direction position X70 (when there is vitreous opacity) and the X-direction position X150 (when there is retinal blood vessel E2) are shown as examples of occurrence of shadows. Shadows also occur when hard vitiligo or the like is present. When the lens opacity exists, a shadow is generated in the tomogram from the inner boundary membrane to the choroid as shown in the tomographic image at the X-direction position X70. When retinal hemorrhage, hard vitiligo, etc. are present, the position in the Z direction where a shadow is generated changes depending on where the retinal hemorrhage or hard vitiligo occurs in the tomography from the nerve fiber layer to the choroid in the Z direction. It will be.

  When a shadow is generated, as described above, motion contrast data may be masked and displayed as an artifact in the OCTA image due to the shadow. In this case, it is difficult to discriminate between the area where artifacts due to shadows appear and the area where there is no temporal change of the subject due to the influence of disease or the like and the motion contrast data is low. There is a risk of lowering. Therefore, in this embodiment, in order to prevent a decrease in the diagnosis accuracy of the doctor, the signal processing unit 300 detects an artifact caused by the shadow.

  Hereinafter, with reference to FIGS. 6 to 10, the processing by the signal processing unit 300 according to the present embodiment will be described in detail. FIG. 6 is a flowchart showing a flow of a series of processes by the signal processing unit 300.

  Note that the following description is a description of processing after data acquisition of the scan area 421 is completed by the imaging optical system 100 and a B scan image is generated by the image generation unit 330 from the acquired data. As described above, the data acquisition of the scan area 421 is performed by the imaging optical system 100 repeating the B scan m times at the Y-direction scan positions from the Y-direction positions Y1 to Yn. In the signal processing unit 300, the image generation unit 330 generates a B scan image based on the interference signal acquired from the imaging optical system 100 by the acquisition unit 310. The generated B scan image is stored in the storage unit 360.

  First, in step S601, the calculation unit 332 of the OCTA image generation unit 331 generates motion contrast data. FIG. 7 is a flowchart of motion contrast data generation processing.

  In step S701, the calculation unit 332 defines an index i at the Y-direction position Yi, and repeats the processing from step S702 to step S707 while the index i is from 1 to n. Thereby, the process from step S702 to step S707 is repeated for the data at each position in the Y direction from the Y position Y1 to Yn.

  In step S <b> 702, the calculation unit 332 reads m B scan images corresponding to the Y-direction position Yi from the storage unit 360. In step S703, the calculation unit 332 corrects the misalignment of the m B scan images read in step S702. The reason for performing the positional deviation correction is that it is necessary to compare B-scan images at the same position at different times in order to generate motion contrast data. The position shift correction is performed by the calculation unit 332 so that the correlation between the B scan images is maximized, for example, while shifting the positions of the B scan images. Note that when the subject does not move like the eye E or when the tracking (tracking) performance is high, the positional deviation correction may not be performed. Further, the positional deviation correction may be performed by a known arbitrary method.

  In step S <b> 704, the arithmetic unit 332 performs an averaging process for each pixel on the m B scan images that have been subjected to the positional deviation correction. As a result, one B-scan image (average B-scan image) obtained by performing the averaging process from the m B-scan images is obtained. In step S <b> 705, the calculation unit 332 stores the average B scan image in the storage unit 360.

  In step S706, the calculation unit 332 calculates motion contrast data using the m B-scan images for which the positional deviation has been corrected in step S703. The motion contrast data is obtained by calculating a dispersion value for each pixel of the m B-scan images for which the positional deviation has been corrected in step S703. Note that the method for calculating motion contrast data is not limited to the method of this embodiment. For example, using the complex OCT signal after the Fourier transform performed by the image generation unit 330 and before the calculation of the absolute value, the motion contrast data is compared with the intensity difference or phase difference between the complex OCT signals with continuous acquisition times. It can also be obtained by calculating the vector difference. In this case, the calculation unit 332 can obtain final motion contrast data as an average value or a maximum value of each intensity difference or the like. The calculation unit 332 calculates the motion contrast data using intensity information (luminance information), phase information, both luminance and phase information, or real part and imaginary part information of the complex OCT signal after Fourier transform. The method may be adopted.

  In step S707, the calculation unit 332 stores the motion contrast data calculated in step S706 in the storage unit 360. In step S708, when the processes from step S702 to S707 are repeated at each position in the Y direction from the Y position Y1 to Yn, the calculation unit 332 ends the motion contrast data generation process.

  Through such motion contrast data generation processing (step S601), the OCTA image generation unit 331 can obtain motion contrast data at each position in the Y direction from the Y position Y1 to Yn.

  Next, in step S602, the calculation unit 332 of the OCTA image generation unit 331 performs a segmentation process on the average B scan image.

  Specifically, first, the calculation unit 332 reads an average B scan image from the storage unit 360. The average B-scan image is stored in association with each position from the Y-direction positions Y1 to Yn, and the calculation unit 332 reads the average B-scan image at all the Y-direction positions. Next, the calculation unit 332 calculates the boundary position of a tomographic layer such as a nerve fiber layer, a ganglion cell layer, and a retinal pigment epithelium for each average B-scan image. The boundary position of the tomography is obtained by performing edge extraction processing in the Z direction on the average B-scan image and analyzing the extracted edge profile. Note that the segmentation processing method is not limited to this, and the calculation unit 332 may specify the boundary position of the fault by any known method. When the boundary position of each slice is obtained for each average B-scan image, the calculation unit 332 stores the boundary position data in the storage unit 360 in association with each average B-scan image.

  Next, in step S603, the mask processing unit 333 of the OCTA image generation unit 331 performs motion contrast data threshold processing (mask processing). FIG. 8 is a flowchart of motion contrast data threshold processing.

  First, in step S801, the mask processing unit 333 defines an index i at the Y-direction position Yi, and repeats the processing from step S802 to step S806 while the index i is from 1 to n. Thereby, the processing from step S802 to step S806 is repeated for the data at each position in the Y direction from Y1 to Yn.

  In step S802, the mask processing unit 333 sets a threshold for threshold processing. The set threshold value is used in the threshold processing of motion contrast data performed in step S804. In this embodiment, the threshold setting value is, for example, a portion where an arbitrary B-scan image or an average B-scan image other than a tomographic image at the Y-direction position Yi is displayed, that is, a portion where only random noise is displayed. The average value of + 2σ. Here, σ represents a standard deviation. Note that the threshold setting value may be arbitrarily set according to a desired configuration for noise reduction. In this embodiment, the threshold value is set for each Y-direction position Yi. However, one threshold value may be set for an area where all Y-direction positions or a plurality of Y-direction positions are combined.

  In step S803, the mask processing unit 333 stores the motion contrast data corresponding to the Y-direction position Yi stored in step S707 and the average B-scan image corresponding to the Y-direction position Yi stored in step S705. Read from.

  In step S804, the mask processing unit 333 performs threshold processing on the motion contrast data. Specifically, the mask processing unit 333 compares the pixel value of the average B scan image with the threshold set in step S802 for each pixel, and if the pixel value of the average B scan image is equal to or less than the threshold, Corresponding motion contrast data is set to 0 (mask). On the other hand, if the pixel value of the average B-scan image is larger than the threshold value, the mask processing unit 333 maintains the value of motion contrast data corresponding to the pixel.

  In step S805, the mask processing unit 333 stores the threshold value-processed motion contrast data in the storage unit 360. Further, in step S806, the mask processing unit 333 causes the storage unit 360 to store the position (specifically, the position in the X, Y, and Z directions) of the pixel in which the value of the motion contrast data is masked by the threshold processing in step S804. . In step S807, when the processes from S802 to S806 are repeated at each position in the Y direction from the Y position Y1 to Yn, the mask processing unit 333 ends the threshold processing of the motion contrast data.

  Through these processes, the OCTA image generation unit 331 can obtain motion contrast data that has been subjected to threshold processing for noise reduction.

  In step S604, the OCTA image generation unit 331 generates an OCTA image based on the motion contrast data on which threshold processing has been performed. Specifically, the OCTA image generation unit 331 reads out the boundary position of each layer obtained in step S602 from the storage unit 360. Next, the OCTA image generation unit 331 extracts motion contrast data corresponding to an arbitrary layer designated by the examiner from the motion contrast data subjected to threshold processing in step S603, using the read boundary position of each layer.

  The OCTA image generation unit 331 determines a representative value in the Z direction of each A scan position for the extracted motion contrast data of the desired layer. The representative value can be an average value, a maximum value, a median value, or the like of motion contrast data of a desired layer corresponding to each A scan position. The OCTA image generation unit 331 generates a two-dimensional OCTA image by generating a representative value for each A scan position in two dimensions in the XY directions. The OCTA image generation unit 331 stores the generated OCTA image in the storage unit 360. The display control unit 370 causes the display unit 400 to display the stored OCTA image.

  In this embodiment, a method for generating an OCTA image for each layer is described. However, the method for specifying motion contrast data used for an OCTA image is not limited to this. Motion contrast data used to generate an OCTA image can be specified for a desired Z-direction range. For example, the range of motion contrast data used for generating the OCTA image may be a range including a plurality of layers, or a position away from the boundary of the layers may be an upper limit or a lower limit of the range. Further, the range may be automatically set by the signal processing unit 300 by designating an examination site or the like.

  Next, with reference to FIG. 9 and FIG. 10, step S605 including the process for detecting the artifact resulting from a shadow is demonstrated. In step S605, the artifact detection unit 350 detects whether or not artifacts due to shadows are displayed in the OCTA image generated in step S804, and the display control unit 370 displays the detection result on the display unit 400. Let FIG. 9 is a flowchart of artifact detection and detection result display processing according to the present embodiment.

  First, in step S901, the shadow specifying unit 340 defines an index i at the Y-direction position Yi, and repeats the processing from step S902 to step S905 while the index i is from 1 to n. Thereby, the processing from step S902 to step S905 is repeated at each position in the Y direction from Y1 to Yn.

  In step S902, the shadow specifying unit 340 reads an average B-scan image corresponding to the Y-direction position Yi from the storage unit 360.

  In step S903, the shadow specifying unit 340 specifies the position of the shadow part from the read average B-scan image. As described above, since the shadow is generated in a tomogram existing behind the shield as viewed from the incident direction of the measurement light, the Z-direction position where the shadow is generated is the position of the shield in the Z direction of the eye E to be examined. It depends on whether it exists. Therefore, in step S903, the shadow specifying unit 340 uses the average B scan image to specify at which position in the X direction the shadow is generated and from which position in the Z direction the shadow is generated behind. Hereinafter, two examples of the method for identifying the shadow portion will be described. In the present embodiment, the shadow specifying unit 340 specifies the X direction position and the Z direction position where the shadow is generated by any one of the two examples.

  The first example is a method of identifying a shadow portion from the luminance value of the average B-scan image. In this method, first, the shadow specifying unit 340 specifies at which position in the X direction the shadow is generated. The shadow specifying unit 340 divides the average B scan image into A scan units, and calculates the sum of all pixel values (luminance values) for each divided A scan unit. The portion where the shadow is generated (shadow portion) has a lower luminance value than the portion where the shadow is not generated. Therefore, the shadow specifying unit 340 can specify the X direction position of the shadow part by specifying the X direction position where the sum of the pixel values corresponding to the A scan is equal to or less than a predetermined threshold.

  Next, the shadow specifying unit 340 specifies from which position in the Z direction the shadow is generated backward. Specifically, the shadow specifying unit 340 corresponds to the pixel value corresponding to the A scan at the X-direction position of the specified shadow portion and the A-scan at the X-direction position where the shadow portion in the vicinity thereof is not detected. The pixel value is compared for each pixel at the same Z direction position from the incident direction of the measurement light. Here, the X direction corresponds to the main scanning direction of the measurement light. Further, the X-direction position where the shadow portion is not detected is the X-direction position around the X-direction position of the specified shadow portion, and the tomographic image appearing at the X-direction position of the specified shadow portion Any position where a similar tomographic image appears may be used. The shadow specifying unit 340 compares the pixel value at the X direction position of the specified shadow part with the pixel value to be compared for each pixel at the same Z direction position from the incident direction of the measurement light, and compares the pixel value with the comparison target pixel value. A Z direction position of a pixel having a value smaller than a predetermined threshold is specified. Accordingly, the shadow specifying unit 340 can detect from which position in the Z direction the shadow is generated rearward.

  It should be noted that by appropriately setting the threshold value for specifying the X-direction position of the shadow portion, it is possible to identify a portion where the luminance value is low due to a lesion or the like and the shadow portion. On the other hand, when specifying the Z direction position of the shadow portion, the shadow portion and the lesion are also compared by similarly comparing the pixel values at the Z direction position after the Z direction position specified by the comparison of the pixel values. Thus, the discrimination accuracy from the low luminance region can be improved. In this case, comparison may be performed for all positions after the specified Z direction position, or comparison may be performed for some of the positions.

  The second example is a method of specifying a shadow portion using the average B-scan image and the segmentation processing result obtained in step S602. In this method, the shadow specifying unit 340 first divides the average B-scan image into each layer using the segmentation processing result. Next, the data of each layer is extracted from the vitreous side in order from the incident direction of the measurement light, in other words, the nerve fiber layer, the ganglion cell layer, and the inner plexiform layer. Further, the shadow specifying unit 340 divides the extracted data of each layer into A scan units. The shadow specifying unit 340 calculates the sum of all pixel values in each layer for each divided A-scan unit. As in the first example, the shadow specifying unit 340 specifies an X-direction position where the result of the sum of pixel values corresponding to the A scan in the data of each layer is equal to or less than a predetermined threshold set for each layer. The shadow specifying unit 340 can specify in which position in the X direction the shadow is generated and from which layer in the Z direction is generated backward by repeating this process for each layer.

  Here, when low luminance is detected in all layers from the nerve fiber layer to the choroid, the shadow specifying unit 340 can specify and detect a shaded part due to vitreous turbidity or lens turbidity. When low luminance is detected from the ganglion cell layer to any one of the boundaries between the inner granule layer and the outer reticulated layer from the incident direction of the measurement light, the shadow specifying unit 340 includes the retina. It is possible to identify and detect a shadow portion due to blood vessels existing in the layer on the incident side of the measurement light. For this reason, when low luminance is detected from a layer where low luminance is detected as viewed from the incident direction of the measurement light to a layer behind it, it can be specified as a shadow portion. At this time, by comparing the pixel value with the threshold value for the layer behind the layer in which the low luminance is detected in the Z direction, it is possible to improve the discrimination accuracy between the shadow portion and the region such as the lesion. it can.

  By such processing, the shadow specifying unit 340 can specify and detect the X-direction position and the Z-direction position of the shadow portion in the tomographic image at the Y-direction position Yi. In step S904, the shadow specifying unit 340 stores the position (X, Y, Z direction position) of the shadow part in the storage unit 360.

  In step S905, the artifact detection unit 350 detects the position of the artifact caused by the shadow in the OCTA image. FIG. 10 is a flowchart showing an artifact position detection process.

  First, in step S <b> 1001, the artifact detection unit 350 reads out the position (X, Y, Z direction position) of the shadow portion stored in step S <b> 904 from the storage unit 360. In step S1002, the artifact detection unit 350 reads the position (X, Y, Z direction position) masked by the threshold processing stored in step S806 from the storage unit 360.

  Next, in step S1003, the artifact detection unit 350 defines an index k at the X direction position Xk, and repeats the processing from step S1004 to step S1007 while the index k is from 1 to p. In step S1004, the artifact detection unit 350 defines an index r at the Z-direction position Zr, and repeats the processing from step S1005 to step S1006 while the index r is from 1 to q.

  In step S1005, the artifact detection unit 350 determines whether or not the pixel position (Xk, Zr) designated in steps S1003 and S1004 corresponds to the position included in the shadow portion and masked by the threshold processing. judge. More specifically, the artifact detection unit 350 reads from the storage unit 360 whether or not the designated position (Xk, Zr) corresponds to the position of the shaded part read from the storage unit 360. It is determined whether it corresponds to the masked position.

  Artifacts due to shadows in the OCTA image are generated when the brightness value of the B-scan image becomes low due to the influence of the shadow caused by the shielding object, and further, the motion contrast data corresponding to the brightness value is masked by threshold processing. . Therefore, the artifact detection unit 350 detects the position of the pixel corresponding to the shadow part and masked by the threshold processing by the determination in step S1005. Thereby, the artifact detection unit 350 can detect the pixel position (Xk, Yi, Zr) of the OCTA image in which the artifact due to the shadow is generated.

  Note that even for a pixel corresponding to a shadow portion, depending on the luminance value of the pixel, the motion contrast data may not be masked by the threshold processing in step S804. For this reason, in the present embodiment, as a condition for specifying an artifact caused by the shadow, it is necessary that the pixel corresponds to the shadow portion and is located at a position where the motion contrast data is masked by the threshold processing.

  In step S1005, if the artifact detection unit 350 determines that the designated pixel corresponds to the shadow portion and is a pixel at the position where the motion contrast data is masked by the threshold processing, the process proceeds to step S1006. In step S1006, the artifact detection unit 350 causes the storage unit 360 to store the designated pixel position as an artifact position (Xk, Yi, Zr) caused by the shadow. If the artifact detection unit 350 determines in step S1005 that the designated pixel is not a shaded part and does not correspond to a pixel at a position masked by threshold processing, the process proceeds to step S1007.

  In step S1007, the artifact detection unit 350 ends the process when the processes of steps S1005 and S1006 are repeated from the Z-direction position Z1 to Zq. In step S1008, the artifact detection unit 350 ends the artifact position detection process when the processes in steps S1004 to S1006 are repeated from the X-direction position X1 to Xp.

  Through the above processing, the artifact detection unit 350 can detect the artifact position in the B-scan image at the Y-direction position Yi. Next, in step S906, the artifact detection unit 350 advances the process to step S907 when the process from S902 to S905 is repeated at each position in the Y direction from the Y direction position Y1 to Yn. As a result of these processes, the artifact detection unit 350 can detect the positions where artifacts due to the shadow are generated at all pixel positions in the X, Y, and Z directions of the three-dimensional motion contrast data.

  Next, in step S907, the display control unit 370 causes the display unit 400 to display an artifact detection result caused by the shadow. At this time, an OCTA image of an arbitrary layer designated by the examiner in step S604 is already displayed on the display unit 400.

  Specifically, first, the artifact detection unit 350 obtains data corresponding to the layer designated by the examiner for generating the OCTA image from the artifact position data obtained in step S905, for each layer identified by the segmentation process. Extract using boundary position. Thereafter, the artifact detection unit 350 determines whether or not the displayed OCTA image includes an artifact due to a shadow, using the extracted data. Then, the display control unit 370 causes the display unit 400 to display the detection result of the artifact caused by the shadow based on the determination of the artifact detection unit 350.

  As an example of the display of the detection result, it may be displayed whether an artifact caused by the shadow is detected or not detected. Further, when an artifact caused by a shadow is detected, an area in which the artifact is detected may be indicated in the OCTA image.

  As a specific example of the detection result, an example in which an artifact is detected in an OCTA image will be described with reference to FIGS. 11 (a) and 11 (b). FIGS. 11A and 11B are examples of OCTA images displayed on the display unit 400. FIG. FIG. 11A shows an OCTA image of the ganglion cell layer, and the presence of retinal blood vessels can be confirmed in the image. On the other hand, FIG. 11B shows an OCTA image of a choriocapillaris plate. FIG. 11B shows regions 1101 and 1102 in which artifacts due to shadows are detected.

  When attention is paid to FIG. 11B which is an OCTA image of the choriocapillaris plate, it can be seen that artifacts due to shadows are generated in the region 1101 and the region 1102. This artifact coincides with the retinal blood vessel shown in FIG. As described above, this is due to the fact that the shadow produced by the retinal blood vessels as a shield exists in the layer of the choriocapillary plate.

  In step S907, the artifact detection unit 350 first extracts data corresponding to the choriocapillary plate layer designated by the examiner from the artifact position data obtained in step S905. Next, the artifact detection unit 350 determines whether the extracted data includes an artifact due to a shadow. Thereby, the artifact detection unit 350 can specify the area where the artifact due to the shadow is detected, that is, the area 1101 and the area 1102 in the example shown in FIG. The display control unit 370 changes the colors of the regions 1101 and 1102 specified with respect to the region where no artifact is detected, or causes the regions 1101 and 1102 to be displayed in a blinking manner, and causes the display unit 400 to display the highlighted regions.

  Through the above processing, the artifact detection unit 350 detects whether or not the artifact is displayed in the OCTA image, the display control unit 370 displays the detection result on the display unit 400, and the process of step S605 is completed. Thereby, a series of processes by the signal processing unit 300 ends.

  As described above, according to the present embodiment, it is possible to detect whether or not artifacts caused by shadows are displayed in the OCTA image. Therefore, whether the low-luminance region of the OCTA image has no temporal change of the subject due to the influence of a disease or the like and the value of motion contrast data is low, or whether the motion contrast data is masked by a shadow and displayed as an artifact. Easy to distinguish. As a result, it is possible to prevent a decrease in the diagnostic accuracy of the doctor based on the presence of artifacts caused by the shadow.

  As described above, the signal processing unit 300 according to the present embodiment includes the acquisition unit 310 that acquires a plurality of tomographic data obtained based on the measurement light controlled to scan the same position of the eye E. The signal processing unit 300 calculates a motion contrast data of the eye E from a plurality of tomographic data, and a mask for masking the motion contrast data based on at least one tomographic data of the plurality of tomographic data. A processing unit 333 is provided. Furthermore, the signal processing unit 300 includes an OCTA image generation unit 331 that generates an OCTA image based on the motion contrast data after the mask processing. Further, the signal processing unit 300 includes a shadow specifying unit 340 that specifies a shadow portion of the eye E based on at least one piece of tomographic data. Furthermore, the signal processing unit 300 includes an artifact detection unit 350 that detects an artifact in the OCTA image based on the identification result of the shadow identification unit 340 and the position where the motion contrast data is masked by the mask processing unit 333.

  Specifically, the shadow specifying unit 340 compares at least one tomographic data of the plurality of tomographic data with a threshold value, and specifies a shadow part based on the comparison result. More specifically, the shadow specifying unit 340 compares tomographic data corresponding to scanning (A scan) in the depth direction of the eye E with the measurement light among the plurality of tomographic data, and based on the comparison result. The position of the shadow portion in the main scanning direction is specified. Further, the shadow specifying unit 340 compares at least one tomographic data of the plurality of tomographic data with the peripheral tomographic data in the main scanning direction by the measurement light, and based on the comparison result, the depth direction of the eye E to be examined The position in (Z direction) is specified. In addition, the shadow specifying unit 340 may compare the tomographic data for each layer of the subject among the plurality of tomographic data with a threshold value and specify the shadow part based on the comparison result.

  According to the signal processing unit 300 according to the present embodiment, it is possible to detect an artifact caused by a shadow in the OCTA image. For this reason, in the OCTA image, it is possible to easily discriminate between an area where the luminance value is low due to the influence of a disease or the like and an area where motion contrast data is masked by a shadow and displayed as an artifact. As a result, it is possible to prevent a decrease in the diagnostic accuracy of the doctor based on the presence of artifacts caused by the shadow.

  In this embodiment, motion contrast data generation processing, mask processing, and shadow specification processing are performed based on the B-scan image. However, these processes may be performed using the tomographic data of the eye E acquired by the acquisition unit 310. As described above, the tomographic data includes the interference signal of the eye E, the signal after the Fourier transform of the interference signal, the signal obtained by subjecting the signal after the Fourier transform to arbitrary signal processing for image generation, and these A signal such as a luminance value based on is included.

  In this embodiment, several examples of display of artifact detection results due to shadows have been shown. However, display examples are not limited to this, and may be realized by other methods as long as the detection results are shown. Good. For example, the detection result of the artifact caused by the shadow may be notified to the examiner by text or voice, or a graphic display and a text or voice notification may be combined. In the present embodiment, the detection result of the artifact caused by the shadow is displayed on the display unit 400. However, the detection result may not be displayed. For example, the detection result of the artifact caused by the shadow may be used for other arbitrary processing.

  In the present embodiment, the shadow specifying unit 340 specifies a shadow part from the average B-scan image corresponding to the Y-direction position Yi in the processing from step S901 to step S903. However, the shadow specifying unit 340 generates an Enface image using each average B-scan image from the Y-direction positions Y1 to Yn and the layer boundary position data obtained in step S602, and the shadow is generated from the Enface image. The part may be specified. Here, the Enface image is a two-dimensional XY direction in which a representative value of each A-scan data is determined from data of an arbitrary layer extracted by the segmentation process, and the determined representative value is generated as a pixel value of each pixel. It is a front image.

  In this case, the shadow specifying unit 340 specifies a position where the pixel value of the Enface image indicates a numerical value equal to or less than a predetermined threshold as the shadow part. Furthermore, the shadow specifying unit 340 specifies the position in the Z direction where the shadow is generated by specifying the shadow part in order from the incident direction of the measurement light for each layer in the specified shadow part. Thereby, the shadow specifying unit 340 can specify at which position in the XY direction the shadow is generated and from which position in the Z direction when viewed from the incident direction of the measurement light. Note that the B scan image used for generating the Enface image may be an arbitrary B scan image out of m sheets at each Y-direction position.

  In this embodiment, three examples of the method for specifying the shadow portion are given. However, the method for specifying the shadow portion is not limited to this, and another method that can detect the shadow portion may be used. Moreover, when specifying the Z direction position of a shadow part, you may compare the value of the pixel for every layer. In this case, the shadow specifying unit 340 obtains at least one tomographic data of the plurality of tomographic data and peripheral tomographic data in the main scanning direction by the measurement light, which indicates the same layer as the layer of the eye E to be examined indicated by the tomographic data. Compare. And the shadow specific | specification part 340 specifies the position in the depth direction of the eye E to be examined based on a comparison result.

  More specifically, the shadow specifying unit 340 obtains the pixel values in each layer at the X-direction position of the specified shadow portion in order of the shadow portions near the X-direction position as viewed from the incident direction of the measurement light. The pixel value of the corresponding depth position of the corresponding layer of the X direction position that has not been detected is compared. For example, the shadow specifying unit 340 includes the top pixel value in the Z direction of the ganglion cell layer in the X direction position of the specified shadow part, and the ganglion cell in the X direction position in which no nearby shadow part is detected. Compare the pixel value at the top of the layer. In this case, since the comparison target is pixels indicating corresponding portions of the same layer, the accuracy of comparison is improved, and the accuracy of specifying the position of the shadow portion in the Z direction is improved.

  In this case, the position in the Z direction for comparing the pixel values is not limited to the position in the order of the incident direction of the measurement light from the top of each layer. For example, the comparison may be made from the pixel value at the center position of each layer. . In this case, in order to specify the generation position in the Z direction of the shadow portion, if the compared pixel value is lower than the threshold value by a threshold value or more than the neighboring comparison target pixel value, the measurement light is incident in the reverse order to the incident direction. If not, the comparison can be continued in the order of the incident direction of the measurement light. Note that the pixel position at the center of each layer can be set to an intermediate position in the Z direction of the boundary position of the layer specified by the segmentation process.

  Further, in the first example, the shadow specifying unit 340 is not limited to the configuration that compares the sum of pixel values corresponding to the A scan with a predetermined threshold when specifying the position of the shadow part in the X direction. For example, an arbitrary statistical value such as an average value, a maximum value, or a median value of pixel values may be compared with a threshold value. Similarly, in the second example, the shadow specifying unit 340 is not limited to the configuration in which the sum of the pixel values in the Z direction in each layer is compared with a predetermined threshold. For example, the average value, the maximum value, or the center of the pixel values Any statistical value such as a value may be compared with a threshold.

  In the present embodiment, detection of artifacts caused by shadows in the OCTA image has been described. However, artifacts due to shadows also occur in motion contrast images based on motion contrast data other than OCTA images. Examples of the motion contrast image other than the OCTA image include a three-dimensional motion contrast image based on three-dimensional motion contrast data, a two-dimensional motion contrast image corresponding to a tomographic image, and the like. The processing for detecting artifacts due to shadows according to the present embodiment can be applied to motion contrast images other than these OCTA images. According to the artifact detection process caused by the shadow according to the present embodiment, it is possible to appropriately detect the artifact caused by the shadow in these images.

  In the present embodiment, the shadow specifying unit 340 specifies the position of the shadow part from the average B scan image, but instead of the average B scan image, an arbitrary B scan image of m sheets, that is, an average process is performed. A B-scan image that is not used may be used. As the B-scan image that has not been averaged, for example, image quality evaluation can be performed using a known method, and the B-scan image with the best image quality can be selected from the m B-scan images. The average B-scan image is less affected by random noise on the image than the B-scan image that has not been averaged. Therefore, when the average B-scan image is used, the average B-scan image is higher when specifying the position of the shadow portion. The accuracy can be specified.

[Example 2]
In the first embodiment, the shadow specifying unit 340 specifies the position of the shadow part from the luminance value of the average B scan image. On the other hand, in Example 2, the shadow specifying unit 340 specifies the position of the shadow part from the motion contrast data generated by the calculation unit 332 of the OCTA image generation unit 331.

  The artifact detection and detection result display processing according to this embodiment will be described below with reference to FIG. Note that the configuration of the OCT apparatus according to the present embodiment is the same as that of the OCT apparatus 50 according to the first embodiment, and therefore the same reference numerals are assigned and description thereof is omitted. Each process according to the second embodiment is the same as that according to the first embodiment except for artifact detection and detection result display processing. Hereinafter, detection of artifacts and display processing of detection results according to the present embodiment will be described focusing on differences from the first embodiment.

  FIG. 12 is a flowchart of artifact detection and detection result display processing according to the present embodiment. In FIG. 12, processes other than steps S1202 and S1203 are the same as the processes according to the first embodiment, and thus description thereof is omitted.

  If the index i of the Y direction position is defined in step S901, the process proceeds to step S1202. In step S1202, the shadow specifying unit 340 reads out the motion contrast data corresponding to the Y-direction position Yi stored in step S707 from the storage unit 360.

  In step S1203, the shadow specifying unit 340 uses the motion contrast data read in step S1202 to determine in which position in the X direction the shadow is generated, and the shadow is in the Z direction as viewed from the incident direction of the measurement light. Specify from which position in the rear the camera originates.

  Hereinafter, the shadow specifying process by the shadow specifying unit 340 according to the present embodiment will be specifically described. As shown in the first embodiment, the motion contrast data obtained in step S601 indicates the position of the blood vessel portion. Further, the shadow is generated at the rear position of the blood vessel portion. For this reason, the shadow specifying unit 340 specifies a blood vessel part by searching and specifying pixels in which the motion contrast data has a numerical value equal to or greater than a predetermined threshold value in the Z direction when viewed from the incident direction of the measurement light. And the shadow specific | specification part 340 specifies the pixel located behind the position of the specified blood-vessel part as a position where a shadow arises. The shadow specifying unit 340 performs the process for each A scan to determine in which position in the X direction the shadow is generated, and from which position in the Z direction the shadow is generated in the backward direction when viewed from the incident direction of the measurement light. Can be detected.

  The shadow specifying unit 340 is a pixel whose motion contrast data has a numerical value equal to or greater than a predetermined threshold, and the luminance value of the pixel of the average B-scan image at the same position is equal to or greater than the threshold set in step S802. The location may be the position of the blood vessel portion. By adding the condition that the luminance value of the pixel of the average B-scan image at the same coordinate position is equal to or greater than the threshold value, the noise component can be removed, and the blood vessel portion can be specified with high accuracy. Instead of the average B scan image, an arbitrary B scan image out of m sheets may be used.

  As described above, in this embodiment, the shadow specifying unit 340 compares the motion contrast data calculated based on a plurality of tomographic data with a threshold value, and specifies a shadow portion based on the comparison result. Even if the method for identifying a shadow part according to the present embodiment is used, the artifact detection unit 350 can detect an artifact due to the shadow based on the position of the specified shadow part and the position masked by the threshold processing. it can. For this reason, the same effect as that of the first embodiment can be obtained by the present embodiment in which the shadow portion is specified using the motion contrast data.

  In the above embodiment, the acquisition unit 310 acquires the interference signal acquired by the imaging optical system 100, the data after Fourier transform generated by the image generation unit 330, and the like. However, the configuration in which the acquisition unit 310 acquires these signals is not limited to this. For example, the acquisition unit 310 may acquire these signals from a server or an imaging device connected to the signal processing unit 300 via a LAN, WAN, or the Internet.

  In the above embodiment, a spectral domain OCT (SD-OCT) apparatus using an SLD as a light source has been described as the OCT apparatus, but the configuration of the OCT apparatus according to the present invention is not limited to this. For example, the present invention can be applied to other arbitrary types of OCT apparatuses such as a time domain OCT apparatus and a wavelength sweep type OCT apparatus. In addition, although the SLO optical system is used as the optical system for capturing the fundus image, the fundus image may be captured using another optical system such as a fundus camera.

  In the above embodiment, a fiber optical system using an optical coupler is used as the dividing means, but a spatial optical system using a collimator and a beam splitter may be used. In addition, the configuration of the imaging optical system 100 is not limited to the above configuration, and a part of the configuration included in the imaging optical system 100 may be configured separately from the imaging optical system 100.

  Furthermore, although the Michelson interference system is used as the interference system, a Mach-Zehnder interference system may be used. For example, a Mach-Zehnder interference system can be used when the light quantity difference is large, and a Michelson interference system can be used when the light quantity difference is relatively small according to the light quantity difference between the measurement light and the reference light.

  In the above embodiment, the eye to be examined is described as the subject. However, the subject is not limited to the subject eye, and may be, for example, skin or an organ. At this time, the present invention can be applied to medical equipment such as an endoscope in addition to the ophthalmologic apparatus. In this case, the segmentation process can be performed according to the structure of the subject.

(Other examples)
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

  While the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. Inventions modified within the scope not departing from the spirit of the present invention and inventions equivalent to the present invention are also included in the present invention. Moreover, each above-mentioned Example and modification can be combined suitably in the range which is not contrary to the meaning of this invention.

300: signal processing unit (information processing apparatus), 310: acquisition unit (acquisition unit), 331: OCTA image generation unit (generation unit), 332: calculation unit (calculation unit), 333: mask processing unit (mask unit), 340: Shadow identification unit (identification unit), 350: Artifact detection unit (detection unit), E: Eye to be examined (subject)

Claims (16)

Acquisition means for acquiring a plurality of tomographic data obtained based on measurement light controlled to scan the same position of the subject;
Calculating means for calculating motion contrast data of the subject from the plurality of tomographic data;
Mask means for masking the motion contrast data based on at least one tomographic data of the plurality of tomographic data;
Generating means for generating a motion contrast image based on the motion contrast data after the mask processing;
Identifying means for identifying a shadow portion of the subject based on at least one tomographic data of the plurality of tomographic data;
Detecting means for detecting artifacts in the motion contrast image based on the identification result of the identifying means and the position where the motion contrast data is masked by the masking means;
An information processing apparatus comprising:   The information processing apparatus according to claim 1, wherein the specifying unit compares at least one tomographic data of the plurality of tomographic data with a threshold, and specifies the shadow portion based on a comparison result.   The specifying means compares the tomographic data corresponding to scanning of the subject in the depth direction by the measurement light among the plurality of tomographic data with the threshold value, and specifies the shadow portion based on a comparison result. The information processing apparatus according to claim 2.   The information according to claim 2 or 3, wherein the specifying unit compares the tomographic data for each layer of the subject with the threshold value among the plurality of tomographic data, and specifies the shadow portion based on a comparison result. Processing equipment.   5. The method according to claim 2, wherein the specifying unit compares the motion contrast data calculated based on the plurality of tomographic data with the threshold value, and specifies the shadow portion based on a comparison result. The information processing apparatus described.   5. The device according to claim 2, wherein the specifying unit compares a pixel value of an Enface image based on at least one of the plurality of tomographic data with the threshold value, and specifies the shadow portion based on a comparison result. The information processing apparatus according to any one of claims.   The specifying unit compares at least one tomographic data of the plurality of tomographic data with peripheral tomographic data in the main scanning direction by the measurement light, and based on the comparison result, the depth of the subject in the shadow portion The information processing apparatus according to claim 1, wherein a position in a direction is specified.   The specifying unit compares at least one tomographic data of the plurality of tomographic data with tomographic data in the main scanning direction by the measurement light, which indicates the same layer as the layer of the subject indicated by the tomographic data. The information processing apparatus according to claim 7, wherein a position of the shadow portion in the depth direction of the subject is specified based on a comparison result. Further comprising display control means for displaying the motion contrast image on a display means;
The information processing apparatus according to claim 1, wherein the display control unit causes the display unit to display an artifact detection result output from the detection unit.   The information processing apparatus according to claim 9, wherein the display control unit causes the display unit to highlight and display a region where the artifact is detected as a detection result of the artifact.   The mask means compares at least one tomographic data of the plurality of tomographic data with a predetermined threshold value, and corresponds to the at least one tomographic data when the at least one tomographic data is less than or equal to the predetermined threshold value. The information processing apparatus according to claim 1, wherein motion contrast data to be masked is masked.   The information processing apparatus according to any one of claims 1 to 11, wherein the subject is an eye to be examined.   The information processing apparatus according to claim 1, wherein the artifact is an artifact caused by a shadow in the motion contrast image.   The information processing apparatus according to claim 1, wherein the artifact is an artifact caused by a shadow of a blood vessel existing in a layer on the incident side of the measurement light. Obtaining a plurality of tomographic data obtained based on measurement light controlled to scan the same position of the subject; and
Calculating motion contrast data from the plurality of tomographic data;
Masking the motion contrast data based on at least one tomographic data of the plurality of tomographic data;
Generating a motion contrast image based on the motion contrast data after the masking;
Identifying a shadow portion of the subject based on at least one tomographic data of the plurality of tomographic data;
Detecting artifacts in the motion contrast image based on a result of specifying the shadow portion and a position where the motion contrast data is masked in the mask processing step;
Including an information processing method.   A program that, when executed by a computer, causes the computer to execute each step of the information processing method according to claim 15.