JP2013017619A - Ophthalmologic image processing apparatus, ophthalmologic image processing method, program for executing the method, and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an apparatus that can generate an image capable of observing the distribution of white spot or the course of blood vessel by using a distance from a base point that is set in a volume space to voxel having a specific value.SOLUTION: An ophthalmologic image processing apparatus includes: a volume image acquisition means 11 for acquiring a volume image of a retina composed of the plurality of voxel: a base point setting means 12 for setting a base point in the volume image; a voxel detection means 13 for searching and detecting specific voxel having a specific value out of the plurality of voxel; a distance calculation means for calculating the distance from the base point to the specific voxel; and an image generation means 15 for generating an image in each display mode depending on the distance.

Description

  The present invention relates to an image generation technique of an ophthalmic image processing apparatus used for ophthalmic medical care, and more specifically, an ophthalmic image processing apparatus, an ophthalmic image processing method, a computer program for causing the computer to execute the method, and the program It relates to the storage medium which memorized.

  With the advent of optical coherence tomography (hereinafter referred to as OCT), a volume image composed of a plurality of two-dimensional tomographic images (hereinafter referred to as tomographic images) of the retina can be obtained. In an ophthalmologic medical site, a user can observe a layer structure, a state and distribution of a lesion, a running of a fundus blood vessel, a temporal change of a retinal layer thickness, and the like from a volume image.

  In order to observe not only OCT but also the state and distribution of a lesion, the running of a blood vessel, and the like from a volume image, a projection image obtained by projecting the luminance value of a voxel on a projection plane perpendicular to the viewing direction is effective. In Patent Document 1, in order to observe the temporal change of the surface shape of the time-series heart volume image, the distance from the boundary surface of the region of interest of the heart in different time phases to the projection surface is calculated along the line-of-sight direction, and the difference Depending on the color, it is displayed. Patent Document 2 discloses a method for calculating the distance between a reference point (center of gravity) of a target tissue and each part on the surface of the target tissue, and displaying the change as a displacement amount in order to diagnose abnormal movement of the target tissue. Has been.

JP2007-228969A Medical image processing apparatus and medical image processing method Patent application title: Ultrasonic diagnostic apparatus

  Highly reflective areas such as retinal vitiligo and blood vessels are observed as high brightness areas in the OCT volume image. In order to observe these high-luminance tissues, a maximum value projection method is effective in which a voxel having the maximum luminance value is projected from the voxels on the line of sight onto a projection plane placed in a three-dimensional space.

  However, the luminance values of the nerve fiber layer located on the vitreous side and the retinal pigment epithelium on the choroid side are as high as those of blood vessels and vitiligo. Therefore, when the maximum value projection method is used, vitiligo and blood vessels are displayed with luminance values similar to those of the nerve fiber layer and retinal pigment epithelium, and are difficult to observe.

  The present invention has been made in view of the above problems, and generates an image that allows observation of vitiligo distribution and blood vessel running using a distance from a base point set in a volume space to a voxel having a predetermined value. With the goal.

  In order to achieve the above object, an ophthalmic image processing apparatus according to an aspect of the present invention has the following arrangement.

  That is, a volume image acquisition unit that acquires a volume image of a retina composed of a plurality of voxels, a base point setting unit that sets a base point in the volume image, and a search for a specific voxel having a predetermined value from the plurality of voxels And voxel detecting means for detecting, distance calculating means for calculating a distance from the base point to the specific voxel, and image generating means for generating an image in a display form corresponding to the distance. .

  According to the present invention, an image having a color corresponding to a distance from a base point to a voxel having a predetermined value as a pixel value is generated. As a result, it is possible to generate an image that allows the user to easily observe the retinal blood vessel running and the vitiligo distribution. Furthermore, it is possible to observe the situation in which the observation object is distributed in the depth direction of the retina from the color information.

1 is a diagram illustrating a functional configuration of an ophthalmologic image processing apparatus 1 according to a first embodiment. The figure which shows the structure of the apparatus connected with the ophthalmic image processing apparatus 1 which concerns on an Example. 5 is a flowchart illustrating a processing procedure of the ophthalmic image processing apparatus 1 according to the first embodiment. The figure which shows the distance of the base point which concerns on Example 1, and the detected voxel Schematic diagram of an image showing blood vessel running and vitiligo distribution according to Example 1 The figure which shows the function structure of the ophthalmic image processing apparatus 1 which concerns on Example 2. FIG. 10 is a flowchart illustrating a processing procedure of the ophthalmic image processing apparatus 1 according to the second embodiment. Schematic diagram showing the layer structure of the retina The figure which shows the distance of the base point when the continuity of IS / OS concerning Example 2 is lost, and the detected voxel The figure which shows the distance of the base point in the case of maintaining the continuity of IS / OS which concerns on Example 2, and the detected voxel

[Example 1]
In the present embodiment, a method for generating an image having a display form corresponding to the distance from the base point set at the upper end of the tomographic image to the voxel having the maximum luminance value will be described. This image allows the user to observe a region having a high luminance value such as blood vessel running or vitiligo distribution. Hereinafter, one aspect of the present embodiment will be described.

  FIG. 2 is a configuration diagram of devices connected to the ophthalmologic image processing apparatus 1 according to the present embodiment. As shown in FIG. 2, the ophthalmic image processing apparatus 1 is connected to the tomographic imaging apparatus 2 and the data server 3 via a local area network (LAN) 4 such as Ethernet (registered trademark). Here, the tomographic imaging apparatus 2 is an apparatus that captures a tomographic image of the eye, and examples thereof include OCT. Since OCT acquires a plurality of tomographic images by one imaging, a volume image of the retina can be acquired by arranging these tomographic images in order. The tomographic imaging apparatus 2 captures a tomographic image of a subject (patient) in accordance with an operation by a user (engineer or doctor), and outputs the acquired volume image to the ophthalmic image processing apparatus 1. Further, the ophthalmic image processing apparatus 1 may be configured to be connected to a data server 3 that stores a volume image obtained by the tomographic imaging apparatus 2 and to acquire a necessary volume image therefrom. Note that these devices may be connected via an interface such as USB or IEEE1394. In addition, the LAN 4 may be connected via an external network such as the Internet.

  FIG. 1 shows a configuration of an ophthalmic image processing apparatus 1 in the present embodiment. The ophthalmic image processing apparatus 1 includes a volume image acquisition unit 11, a base point setting unit 12, a voxel detection unit 13, a distance calculation unit 14, and an image generation unit 15.

  Next, the processing procedure of the ophthalmic image processing apparatus 1 in the present embodiment will be described using the flowchart of FIG.

  In step S <b> 301, the volume image acquisition unit 11 acquires a volume image captured by the tomographic imaging apparatus 2 or a volume image stored in the data server 3 and outputs the volume image to the voxel detection unit 13. The volume image is composed of a plurality of voxels.

  In step S <b> 302, the base point setting unit 12 sets a base point of distance and outputs the distance to the distance calculation unit 14. In this embodiment, the base point 401 is set on the upper end of the tomographic image 4 shown in FIG. 4, that is, on the XY plane at Z = 0. As shown in other embodiments to be described later, the XY plane as the starting position is the case of this embodiment, and the base point is a point according to an arbitrary condition in the acquired volume image. It can be set.

  In step S303, the voxel detection unit 13 searches for a voxel 402 having a predetermined value as a specific voxel while scanning the voxel from the base point in the depth direction of the retina along the Z axis of the tomographic image 4 illustrated in FIG. To detect. Here, the voxel having a predetermined value is, for example, a voxel having the maximum luminance value among a plurality of voxels scanned in the depth direction. This is because the blood vessels and vitiligo to be observed are depicted as high-intensity tissue in the tomographic image captured by OCT. On the other hand, the luminance value of the tissue between IS / OS (visual cell inner and outer segment junction) 405 and RPE (retinal pigment epithelium) 406 shown in FIG. However, since light does not reach IS / OS 405 in a region where blood vessels or vitiligo are present, the tissue between IS / OS 405 and RPE 406 is depicted with a low luminance value. Therefore, the voxel detection unit 13 detects a voxel between IS / OS and RPE as a maximum luminance value voxel in a region where there are no blood vessels or vitiligo, and a voxel between blood vessels or vitiligo in a region where they exist. The coordinates of the detected voxels are output to the distance calculation unit 14.

  In step S304, the distance calculation unit 14 calculates the difference between the coordinates of the base point 401 set in step S302 and the coordinates of the specific voxel 402 detected in step S303, and calculates the distance 403 in the tomographic image 4 shown in FIG. Ask. The obtained distance is output to the image generation unit 15.

  In step S305, a pixel value of a display form corresponding to the distance obtained in step S304 is obtained. The image generation unit 15 sets a display form used in actual image display according to the distance. For example, when a color is used for the display form, a color table is used to set a color according to the distance. If the color table is set so that the color changes linearly according to the distance, it is possible to read the depth information in which the maximum luminance value voxel exists from the color information of the image. In addition, shading may be used for the display form. In this case, if the color is set lighter as the distance is longer and darker, the depth information can be read from the light and dark information of the image.

  In step S306, it is confirmed whether or not the base point has been set for all the voxels on the XY plane for which the first base point has been set. If the setting has not been completed, the process returns to step S302. The processing from step S302 to step S305 is repeated to detect the maximum luminance value voxel along the Z axis while setting all the voxels on the XY plane with Z = 0 as the base point, and according to the distance between the two voxels. Set color to pixel value. Accordingly, a plurality of maximum luminance value voxels are obtained according to the number of base points. In addition, the color tone, shading, etc. are set according to the display mode set as described above.

  If it is recognized that the base point setting for all the voxels on the XY plane and the display mode setting based on the voxels have been completed, the process proceeds to step S307. In step S307, all the set display forms are matched to generate a new volume image. In this way, an image is generated by projecting the distance information on a plane parallel to the XY plane of the tomographic image 4 in FIG.

  According to the configuration described above, an image having a pixel value as a color corresponding to the distance from the set base point to the maximum luminance value voxel is generated. The Z coordinate where the maximum luminance value voxel is detected is different between a region where blood vessels and vitiligo are present and a region where blood vessels and vitiligo are not present. Therefore, as shown in FIG. 5, images with different gradation values depending on the distance are generated, and the positions of the vitiligo 501 and the blood vessel 502 can be identified.

(Modification 1-1)
In the above embodiment, the predetermined value is set as the maximum luminance value in step S303. In this modification, a case where the value is equal to or greater than a set threshold value will be described.

  In step S303, the voxel detection unit 13 sets a threshold regarding the luminance of the voxel, detects one or more voxels having a luminance value equal to or higher than the threshold, and calculates, for example, an average value of those Z coordinates. The calculated value is output to the distance calculation unit 14. Accordingly, there are a plurality of specific voxels detected in this case, and the distance calculation is performed a plurality of times between the base point and the coordinates of the voxels having these characteristics.

  In step S <b> 304, the distance calculation unit 14 calculates the difference between the value calculated in step S <b> 303 and the coordinates of the base point as a distance, and outputs the distance to the image generation unit 15.

  By carrying out in this way, it is possible to suppress the influence of high-intensity noise included in the tomographic image of OCT, or to indicate the distance to the average value of the Z coordinates of the high-intensity region having a thickness. .

[Example 2]
In the first embodiment, the method of generating an image that can observe blood vessel running and vitiligo distribution in the case where the base point is a voxel on the XY plane with Z = 0 has been described. In the present embodiment, a method for a case where the base point is a voxel on the layer boundary of the retina will be described.

  FIG. 6 shows the configuration of the ophthalmic image processing apparatus 1 in the present embodiment. The ophthalmologic image processing apparatus 1 includes a volume image acquisition unit 61, a base point setting unit 62, a voxel detection unit 63, a distance calculation unit 64, an image generation unit 65, and a layer boundary detection unit 66.

  Next, the processing procedure of the ophthalmic image processing apparatus 1 in the present embodiment will be described using the flowchart of FIG.

  In step S <b> 701, the volume image acquisition unit 61 acquires a volume image captured by the tomographic imaging apparatus 2 or a volume image stored in the data server 3, and outputs the volume image to the voxel detection unit 63 and the layer boundary detection unit 66. To do.

  In step S702, the layer boundary detection unit 66 extracts each layer inside the retinal layer from the volume image. The layer extraction inside the retina layer will be described with reference to FIG. FIG. 8 shows each of the layers 801 to 810 in the retinal layer and one pixel column A (A-scan) parallel to the z-axis direction. 801 is ILM (inner boundary membrane), 802 is NFL (nerve fiber layer), 803 is GCL (ganglion cell layer), 804 is IPL (inner plexus layer), 805 is INL (inner granule layer), 806 is OPL ( Outer reticulated layer), 807 represents ONL (outer granule layer), 808 represents ELM (outer boundary membrane), 809 represents IS / OS (inner joint of outer segment of photoreceptor cell), and 810 represents RPE (retinal pigment epithelium layer) . Further, the boundary between layers is expressed as “upper layer / lower layer”. In the present embodiment, the layer boundary detection unit 66 extracts at least the ILM 801 and the NFL 802 / GCL 803, the IPL 804 / INL 805, the OPL 806 / ONL 807, the IS / OS 809, and the RPE 810. Note that the extraction of the layer by the layer boundary detection unit 66 is the same as the extraction of the boundary between layers. For example, the NFL 802 is extracted by extracting the boundary between the ILM 801 at the boundary between the vitreous body and the retina and the NFL 802 / GCL 803.

  Regarding the extraction of the retinal layer, first, an image is created by applying a median filter and a Sobel filter to a tomographic image (hereinafter referred to as a median image and a Sobel image). Next, a profile is created for each A-scan from the converted median image and Sobel image. The median image has a luminance value profile, and the Sobel image has a gradient profile. Then, a peak in the profile created from the Sobel image is detected. The retinal layer is detected by referring to the median image profile before and after the detected peak and between the peaks. The detected layer boundary information is output to the base point setting unit 62 and the voxel detection unit 63.

  In step S703, the base point setting unit 62 confirms the continuity of IS / OS. As shown in FIG. 9, in the tomographic image 9 of OCT, the luminance value of the voxel at the back becomes darker due to the presence of the high luminance region 907. Further, IS / OS is not drawn in the optic nerve head. Therefore, the IS / OS 905 is divided as shown in FIG. 9 in the volume image in which the high luminance area exists and the volume image of the optic nerve head. Therefore, the base point setting unit 62 confirms the continuity of the IS / OS by using the input layer boundary information. For example, when the layer boundary is expressed by a point sequence of three-dimensional coordinates, it is determined that continuity is lost when adjacent points are separated by a certain distance or more.

  If the continuity of IS / OS is lost, in step S704, an ILM existing in many OCT tomographic images is set as a base point. For example, as shown in FIG. 9, a base point 901 on the ILM 904 is used. By projecting the ILM onto the XY plane and flattening the ILM, it is possible to generate an image in which the influence of the unevenness of the ILM is reduced.

  In step S706, the voxel detector 63 detects the voxel 902 having the maximum luminance value while scanning the voxel in the depth direction along the Z axis of the tomographic image 9 shown in FIG.

  In step S707, the distance calculation unit 64 calculates the difference between the coordinates of the base point set in step S704 and the coordinates of the voxels detected in step S706, and obtains the distance 903 in the tomographic image 9 shown in FIG. The obtained distance is output to the image generation unit 65.

  In step S708, a color corresponding to the distance 903 obtained in step S707 is set as a pixel value. Further, a display form is set according to the obtained distance 903. In the OCT tomogram, blood vessels are distributed more in the interior of the retina than in the ILM. Therefore, for example, the color table is designed to be red as the distance is short, black as the distance is long, and white in the middle. By designing in this way, it is possible to express blood vessels existing in the ILM in red, white spots in the retina white, and IS / OS that is not an observation target in black.

  In step S709, it is confirmed whether or not the base point has been set for all voxels in the region where the base point can be set on the layer boundary where the base point has been set. If the setting has not been completed, the process returns to step S703.

  The processing from step S703 to step S709 is repeated to detect the maximum luminance value voxel along the Z axis while setting all the voxels on the ILM as the base point, and the color corresponding to the distance between the two voxels is set to the pixel value. Set.

  If it is recognized that the base point setting for all the voxels on the layer boundary and the display mode setting based on the voxels have been completed, the process proceeds to step S710. In step S710, all the set display forms are matched to generate a new volume image. In this manner, an image is generated by projecting the distance information on a plane parallel to the XY plane of the tomographic image 9 in FIG.

  If IS / OS continuity is maintained, in step S705, IS / OS has less undulations than ILM, so the voxel on IS / OS is set as a base point. For example, as shown in FIG. 10, a base point 1001 on the IS / OS 1005 is used. When the entire retina is curved, the IS / OS is projected onto the XY plane and the IS / OS is flattened to generate an image in which the influence of the retina curvature is reduced.

  In step S706, the voxel detection unit 63 detects the voxel 1002 having the maximum luminance value while scanning the voxels existing between the ILM 1004 and the IS / OS 1005 along the Z axis of the tomographic image 10 illustrated in FIG. .

  In step S707, the distance calculation unit 64 calculates the difference between the coordinates of the base point set in step S705 and the coordinates of the voxel detected in step S706, and obtains the distance 1003 in the tomographic image 10 shown in FIG. The obtained distance is output to the image generation unit 65.

  In step S708, a color corresponding to the distance obtained in step S707 is set as a pixel value. Further, a display form is set according to the obtained distance 903. The color table for setting the color according to the distance is designed, for example, red as the distance is long, black as the distance is short, and white in the middle. By designing in this way, it is possible to express blood vessels red, white spots white, and IS / OS black.

  In step S709, it is confirmed whether or not the base point has been set for all voxels in the region where the base point can be set on the layer boundary where the base point has been set. If the setting has not been completed, the process returns to step S703.

  The processing from step S703 to step S709 is repeated to detect the maximum luminance value voxel along the Z axis while setting all the voxels on the IS / OS as the base point, and set the color according to the distance between the two voxels. To do.

  If it is recognized that the base point setting for all the voxels on the layer boundary and the display mode setting based on the voxels have been completed, the process proceeds to step S710. In step S710, all the set display forms are matched to generate a new volume image. In this way, an image is generated by projecting the distance information on a plane parallel to the XY plane of the tomographic image 10 in FIG.

  By using the layer boundary information, it is possible to automatically design a color table in which colors are set for the layer boundaries. Further, it is not affected by the luminance value outside the retina. Therefore, it is possible to generate a more easily viewable image showing the blood vessel running and the vitiligo distribution in the retina.

(Other examples)
The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, or the like) of the system or apparatus reads the program. It is a process to be executed.

DESCRIPTION OF SYMBOLS 1 Ophthalmological image processing apparatus 11 Volume image acquisition part 12 Base point setting part 13 Voxel detection part 14 Distance calculation part 15 Image generation part

Claims (9)

Volume image acquisition means for acquiring a volume image of the retina composed of a plurality of voxels;
Base point setting means for setting a base point in the volume image;
Voxel detection means for searching for and detecting a specific voxel having a predetermined value from the plurality of voxels;
Distance calculating means for calculating a distance from the base point to the specific voxel;
Image generating means for generating an image of a display form according to the distance;
An ophthalmologic image processing apparatus comprising: The predetermined value is a maximum luminance value,
The ophthalmic image processing apparatus according to claim 1, wherein the voxel detection unit searches for and detects a voxel having the maximum luminance value in a depth direction of the retina from the base point. The voxel detection means includes threshold setting means for setting a threshold,
The ophthalmic image processing apparatus according to claim 1, wherein the predetermined value is a luminance value equal to or greater than the threshold value. There are a plurality of the specific voxels,
4. The ophthalmic image according to claim 1, wherein the distance calculation unit calculates the distance from specific coordinates obtained based on the coordinates of each of the plurality of voxels and the base point. 5. Processing equipment. A layer boundary detecting means for detecting a boundary between at least the inner granule layer and the photoreceptor inner / outer segment junction among the layer boundaries of the retina;
The base point setting means sets the base point on either the inner granule layer or the photoreceptor inner-node outer joint joint according to the continuity of the inner-node outer joint joint of the photoreceptor cell. The ophthalmic image processing apparatus according to claim 1, wherein the ophthalmic image processing apparatus is characterized.   The ophthalmic image processing apparatus according to claim 1, wherein the display form is at least one of a form using a change in color and a form using a change in shading. A volume image acquisition step of acquiring a volume image of the retina composed of a plurality of voxels;
A base point setting step for setting a base point of distance in the volume image;
A voxel detection step of searching for and detecting a specific voxel having a predetermined value from the plurality of voxels;
A distance calculating step of calculating a distance from the base point to the specific voxel;
An image generation step of generating an image in a display form according to the distance;
An ophthalmologic image processing method comprising:   A computer program for causing a computer to execute the ophthalmic image processing method according to claim 7.   A storage medium storing the computer program according to claim 8.

Images