Fundus image display apparatus, control method thereof, and computer program

The present invention relates to a fundus image display device, a control method thereof, and a computer program.

In the field of ophthalmology, a fundus camera for taking a photograph of the fundus has been generally used. In recent years, the advent of optical coherence tomography (OCT) has made it possible to take tomographic images of the retina. As a result, new findings in ophthalmic medicine are being discovered, such as the fact that the retina forms a layered structure and that the layered structure collapses as the disease progresses. Currently, as disclosed in Patent Document 1, a layered structure is extracted from retinal volume data reconstructed from multiple tomographic images of the retina taken, and the information is used for diagnosing ophthalmic diseases. I'm doing it. Hereinafter, the fundus photograph and the retinal volume data will be referred to as a fundus image.

Conventionally, ophthalmologists use tomographic images and volume data of the retina to interpret the layered structure of the retina, and fundus photographs to interpret the state of fundus blood vessels and white spots. In particular, the fundus blood vessel is the only blood vessel that can be observed from outside the body, and signs and phenomena of various diseases can be confirmed from the fundus blood vessel. Among them, the crossing phenomenon in which a blood vessel blocked by arteriosclerosis of the fundus blood vessel expands and comes into contact with a nearby blood vessel is known as a phenomenon that causes blood vessel rupture or the like and may lead to blindness in the worst case. Therefore, it can be said that it is meaningful for an ophthalmologist to know the running of the fundus blood vessels in order to make a diagnosis. Furthermore, for patients, it leads to early detection of illness, and early recovery and avoidance of blindness can be expected. For this reason, Patent Document 2 discloses a method of extracting a two-dimensional blood vessel region from a fundus photograph or an integrated image generated by integrating brightness values in the depth direction. Further, as a method for extracting a blood vessel region and a vitiligo region from a fundus photograph, methods such as Non-Patent Document 1 and Non-Patent Document 2 are disclosed, respectively.

However, in the fundus photograph, only the two-dimensional running of the fundus blood vessels could be observed. Therefore, it was not possible to directly observe the three-dimensional overlap of fundus blood vessels seen in the crossover phenomenon. Now that OCT has appeared and retinal volume data can be reconstructed from high-resolution tomographic images, it is possible to observe the three-dimensional running of fundus blood vessels and directly observe the crossover phenomenon.

In order to observe the retinal volume data, a method called volume rendering, which can display the volume data semi-transparently by converting the value of the voxel into opacity or color by a transfer function, is effective. As shown in FIG. 1, the transfer function that influences transparency is expressed as a function in which the luminance value is defined on the horizontal axis and the opacity is defined on the vertical axis, for example. Generally, the user can manually set the shape of the transfer function, the position of the peak, and the width by the user interface. Also, as in Patent Document 3, using the fact that the CT value histogram of organs and blood vessels reflected in CT shows peaks for each organ, a Gaussian function is fitted to the CT value histogram, and CT calculated from the mean value and variance. It can be automatically designed to make the range of values opaque.<patcit num="1"><text>Japanese Unexamined Patent Publication No. 2008-073099</text></patcit><patcit num="2"><text>Japanese Patent Application Laid-Open No. 2007-325831</text></patcit><patcit num="3"><text>Japanese Patent Application Laid-Open No. 2008-6274</text></patcit><nplcit num="1"><text>Elisa Ricci, Renzo Perfetti, "Retinal Vascular Segmentation Using Line Operator and Support Vector Classification", IEEE Transactions, Medical Imaging, Vol. 26, No. 10, pp. 1357-1365, 2007 (Elisa Ricci, Renzo) Perfetti, "Retinal Blood Vessel Segmentation Using Line Operators and Support Vector Classification," IEEE Transactions on Medical Imaging, Vol.26, No.10, pp.1357-1365, 2007.)</text></nplcit><nplcit num="2"><text>Thomas Walter, Jean Claude Klein, Pascal Massin, Ali Elginy, "Image Processing Assistance for Diabetic Retinopathy Diagnosis" IEEE Transactions, Medical Imaging, Vol. 21, No. 10, pp. 1236-1243, 2002 October (Thomas Walter, Jean-Claude Klein, Pascale Massin and Ali Erginay: "A Contribution of Image Processing to the Diagnosis of Diabetic Retinopathy --Detection of Exudates in Color Fundus Images of the Human Retina," IEEE Transactions on Medical Imaging, Vol.21, No.10, pp.1236-pp.1243, Oct. 2002.)</text></nplcit>

<p> In order to visualize the three-dimensional running of the fundus blood vessels using the retinal volume data, there are the following problems.</p><p> The method according to Patent Document 2 and Non-Patent Document 1 is a method of extracting a fundus blood vessel of a two-dimensional fundus photograph or an integrated image, and the three-dimensional position of the fundus blood vessel cannot be specified.</p><p> In the method according to Patent Document 3, since the blood vessel region does not show peaks in the OCT luminance value histogram, it is difficult to automatically visualize the three-dimensional running of the fundus blood vessels. Further, in the retinal tomographic image 201 shown in FIG. 2, the vicinity of the lower end of the nerve fiber layer in which the fundus blood vessel 202 runs is a high-intensity region. Further, the fundus blood vessel 202 reflects light well and the brightness value becomes high. That is, the area in which the fundus blood vessel 202 travels has low contrast, and it is difficult to visualize only the fundus blood vessel even if the transfer function is manually set based on the brightness value.</p><p> The present invention has been made in view of the above problems, and an object of the present invention is to make it possible to visualize three-dimensional running of blood vessels, three-dimensional distribution of vitiligo, etc. from retinal volume data.</p>

<p> The present invention for achieving the above object is a fundus image display device. A specific means for identifying the boundary position of any of the retinal layers in a fundus image showing a tomographic image of the retina, and A distance transfer function setting means for setting a distance transfer function for converting a distance from the specified boundary position into a parameter representing opacity so that the peak position of opacity is a predetermined position in the retina. Luminance transfer function setting means for setting a luminance transfer function that converts the luminance value of the fundus image into a parameter representing opacity, and A visualization means that calculates the opacity of each position of the tomographic image using the distance transfer function and the brightness transfer function, and generates a translucent display image by volume rendering. To be equipped.</p>

<p> According to the configuration of the present invention, it is possible to visualize the three-dimensional running of blood vessels, the three-dimensional distribution of vitiligo, etc. from the retinal volume data.</p>

Hereinafter, preferred embodiments of the fundus image display device according to the present invention will be described in detail with reference to the accompanying drawings. However, the scope of the invention is not limited to the illustrated examples.

[First Embodiment: Visualization of blood vessels] FIG. 3 is a diagram showing an example of the configuration of the fundus image display system according to the present embodiment. In the present embodiment, the fundus image display device 1 can read the fundus image from the database 2 via LAN3. Alternatively, a storage device such as an FDD, a CD-RW drive, an MO drive, a ZIP drive, or the like may be connected to the fundus image display device 1 and the fundus image may be read from those drives. Further, a medical image or the like may be acquired directly from the fundus imaging device 4 via LAN 3. In addition to the fundus image, database 2 stores patient names and findings, and thickness information of the retina 403, nerve fiber layer 405, outer plexiform layer 406, and retinal pigment epithelium 407 shown in FIG. FIG. 4 is a diagram for explaining the relationship between the fundus photograph, the retinal volume data, and the integrated image.

Examples of the fundus imaging device 4 include an OCT capable of capturing a tomographic image of the fundus and a fundus camera capable of capturing a photograph of the fundus. Examples of the type of OCT include TD-OCT of the time domain method and FD-OCT of the Fourier domain method. Generally, in TD-OCT, one tomographic image is taken. On the other hand, in FD-OCT capable of high-speed imaging, a plurality of tomographic images can be obtained by one imaging, and retinal volume data can be reconstructed by arranging these tomographic images in order. The fundus imaging device 4 captures a fundus image of a subject (patient) in response to an operation by a user (engineer or doctor), and outputs the obtained fundus image to the fundus image display device 1. Further, the fundus image display device 1 may be configured to be connected to a database 2 for storing a fundus image or the like obtained by the fundus image capturing device 4 and to acquire a necessary fundus image or the like from the database 2. The connection with these devices may be performed via an interface such as USB or IEEE1394. Further, it may be configured to be connected via an external network such as the Internet.

Next, the functional configuration of the fundus image display device 1 will be described in the present embodiment with reference to FIG. FIG. 5 is a diagram showing an example of the functional configuration of the fundus image display device 1 corresponding to the embodiment of the invention. In FIG. 5, the fundus image display device 1 includes a fundus image input unit 501, a layer extraction unit 502, a blood vessel region extraction unit 503, a distance calculation unit 504, a distance transfer function setting unit 505, a brightness transfer function setting unit 506, and a visualization unit 507. To be equipped. Next, the operation of each part will be described.

The fundus image input unit 501 selects and inputs the fundus image output from the fundus image capturing device 4 and the fundus image stored in the database 2. FIG. 4 shows a schematic diagram of the retinal volume data 401 and the fundus photograph 421 input as the fundus image. It is medically known that the retina has a layered structure. The coordinate system of the retinal volume data 401 has an X-axis in the horizontal direction of the retinal tomographic image 402 (schematic diagram), a Z-axis in the vertical direction, and a Y-axis in the direction in which the retinal tomographic images 402 are arranged. The coordinate system of the fundus photograph 421 has the X-axis in the horizontal direction and the Y-axis in the vertical direction.

The layer extraction unit 502 extracts the nerve fiber layer 405, the outer plexiform layer 406, and the retinal pigment epithelium 407 from the retinal volume data 401. Then, the retina 403 defined by the inner limiting membrane 404, the lower end of the nerve fiber layer 405, the upper end of the outer plexiform layer 406, the upper end of the retinal pigment epithelium 407, and the inner limiting membrane 404 to the lower end of the retinal pigment epithelium 407 is obtained. The layers to be extracted may be all layers, or one or a plurality of layers may be selected. The layer boundary information indicating the boundary position of the extracted layer is output to the distance calculation unit 504.

The blood vessel region extraction unit 503 extracts the blood vessel region from the input fundus image. As shown in FIG. 6, the blood vessel region extraction unit 503 is composed of an integrated image blood vessel region extraction unit 601, a fundus photographic blood vessel region extraction unit 602, and a blood vessel alignment unit 603. The integrated image blood vessel region extraction unit 601 extracts the blood vessel region 412 of the fundus from the integrated image 411 obtained by integrating the voxel brightness values of the retinal volume data 401 in the z-axis direction. Further, when the fundus photograph is input by the fundus image input unit 501, the fundus photographic blood vessel region extraction unit 602 extracts the blood vessel region 422 from the fundus photograph 421. When the fundus photograph is input in the fundus image input unit 501, the blood vessel alignment unit 603 uses the blood vessel region 412 and the blood vessel region 422 to form the coordinate system (X, Y) of the fundus photograph and the coordinate system (x) of the integrated image. , y) is aligned.

As a result of these processes, the blood vessel region extraction unit 503 represents the blood vessel region as a set of voxels represented by the coordinate system (x, y) as a distance calculation unit 504, a distance transfer function setting unit 505, and a brightness transfer function setting unit 506. Output to.

The distance calculation unit 504 first identifies one layer boundary from the layer boundary information, and sets the boundary position to distance = 0. For example, when the internal limiting membrane is specified by selection, the distance of all voxels on the layer boundary extracted as the internal limiting membrane 404 shown in FIG. 4 is set to 0. Next, the distance calculation unit 504 uses the coordinates (x, y) of the retinal volume data using the blood vessel region 412, which is a set of voxels represented by the coordinate system (x, y) obtained by the blood vessel region extraction unit 503. Defines a vascular projection region 408, which is a collection of voxels with). Further, the distance calculation unit 504 calculates the distance between each voxel in the blood vessel projection region 408 and the reference 409 existing at the intersection of the blood vessel projection region 408 and the layer boundary, and gives the calculated distance to each voxel. At this time, the sign of the distance of the voxel having the coordinates smaller than the reference 409 is "-", and the sign of the distance of the voxel having the coordinates larger than the reference 409 is "+".

The distance transfer function setting unit 505 sets a distance transfer function that converts the distance of each voxel calculated by the distance calculation unit 504 into opacity. The distance transfer function is, for example, a function in which distance is defined on the horizontal axis and opacity is defined on the vertical axis in FIG. 1, and is automatically set according to a display target such as a blood vessel or vitiligo. The distance transfer function setting unit 505 sets the distance from the reference 409 to the fundus blood vessel 202 where the fundus blood vessel is considered to travel in the blood vessel projection region 408 extracted by the blood vessel region extraction unit 503 as the peak position of the distance transfer function.

Here, the position where the fundus blood vessel 202 is considered to pass strongly reflects light, and the brightness value of the shadow region 203 of the blood vessel shown in FIG. 2 is lower than that of the surroundings. Therefore, a voxel having the maximum brightness value in the blood vessel projection area 408 or a rectangle of an arbitrary size is set and used as the z coordinate of the voxel in which the variance in the rectangle becomes large. The variance of the distance transfer function can also be set from the blood vessel diameter that can be calculated from the blood vessel region. The distance transfer function automatically set in this way is output to the visualization unit 507.

The brightness transfer function setting unit 506 sets a brightness transfer function that converts the brightness value of the retinal volume data into opacity. The luminance transfer function is, for example, a function in which the luminance value is defined on the horizontal axis and the opacity is defined on the vertical axis in FIG. This function may be set manually by the user using the user interface, or may be automatically set from the histogram of the luminance value as in Patent Document 3. The set luminance transfer function is output to the visualization unit 507.

The visualization unit 507 combines the opacity calculated by the brightness transfer function and the opacity calculated by the distance transfer function by the following equation 1 and uses it to set the opacity of the voxel at the time of volume rendering. Here, the brightness value of the voxel at the coordinate x is v (x), the distance is d (x), and the brightness transfer function is f._v(v (x)), distance transfer function f_dLet it be (d (x)). The combined opacity α (v (x), d (x)) is the linearity of the luminance transfer function and the distance transfer function using the composite ratio β of the luminance transfer function and the distance transfer function, as shown in Equation 1. Obtained by sum.

α (v (x), d (x)) = β * f_v(v (x)) + (1-β) * f_d(d (x)) (Equation 1) Note that β may be set in advance for each display target. When it is desired to visualize a blood vessel, since the contrast is low near the blood vessel, it is conceivable to set β low and increase the weight of the distance transfer function. Finally, the visualization unit 507 generates a semi-transparent display image as the result image of the volume rendering.

Next, the processing procedure for visualizing the three-dimensional running of the fundus blood vessels in the fundus image display device 1 of the present embodiment described above by volume rendering will be described with reference to the flowchart shown in FIG. The process shown by the flowchart of FIG. 8 is realized by executing the program in which the CPU 5 shown in FIG. 12 is stored in the main memory 6.

In step S801, the fundus image input unit 501 inputs the fundus image such as the retinal volume data and the fundus photograph to the fundus image display device 1. The input fundus image is output to the layer extraction unit 502, the blood vessel region extraction unit 503, the distance transfer function setting unit 505, the brightness transfer function setting unit 506, and the visualization unit 507.

In step S802, the blood vessel region extraction unit 503 extracts the blood vessel regions 412 and 422 from the retinal volume data and the fundus image. The process in this step will be described in more detail later with reference to FIG.

In step S803, the layer extraction unit 502 extracts the nerve fiber layer 405, the outer plexiform layer 406, and the retinal pigment epithelium 407 from the retinal volume data received in step S801. The layer extraction unit 502 obtains the inner limiting membrane 404, the lower end of the nerve fiber layer 405, the upper end of the outer plexiform layer 406, and the retina 403 from the information of these layers. The user selects the layer boundary to be used as a reference by the instruction unit (not shown), and the layer boundary information is output to step S804.

In step S804, the distance calculation unit 504 calculates the distance from the layer boundary to each voxel in the z-axis direction within the blood vessel region extracted in step 802 with reference to the selected layer boundary. The calculated distance information is output to the visualization unit 507. When the distance to the voxel other than the blood vessel region is not calculated, that is, when the distance is 0, the opacity of the voxel other than the blood vessel region becomes 0, and the volume rendering can be calculated at high speed. Further, when calculating the distance to the voxel other than the blood vessel region, it is possible to generate a translucent display image by volume rendering while calculating the opacity according to the distance from the layer.

In step S805, the luminance transfer function setting unit 506 and the distance transfer function setting unit 505 set the luminance transfer function and the distance transfer function, respectively. The distance transfer function setting unit 505 extracts the shadow region of the blood vessel, and sets the distance from the inner limiting membrane to the coordinates of a predetermined position where the blood vessel is considered to travel as the peak position. The set luminance transfer function and distance transfer function are output to the visualization unit 507.

In step S806, the visualization unit 507 performs volume rendering while multiplying the opacity obtained from the luminance transfer function and the distance transfer function by Equation 1.

FIG. 9 is a flowchart illustrating the processing in step S802 in detail. First, in step S901, the integrated image blood vessel region extraction unit 601 generates the integrated image 411 of the input retinal volume data 401. Further, the integrated image blood vessel region extraction unit 601 extracts the blood vessel region 412 from the integrated image 411 and outputs it to the blood vessel alignment unit 603 in step S902.

Here, as a method for extracting the blood vessel region 412 in step S902, any known method can be used. For example, by analyzing the pixel values of the integrated image, calculating the difference between the pixel values of the adjacent pixels, and searching for the adjacent pixels whose difference is larger than the predetermined value, the boundary between the blood vessel region and the other region. Detect the area. Thereby, the blood vessel region can be extracted from the integrated image. In this extraction process, the difference in pixel values (luminance value, etc.) between the blood vessel region and other regions in the integrated image is used.

In step S903, the fundus image input unit 501 determines whether or not the fundus photograph 421 is input in step S801. If the fundus photograph 421 is input (YES in S903), the process proceeds to step S904. If the fundus photograph 421 is not entered (NO in S903), the process proceeds to step S906.

In step S904, the fundus photograph blood vessel region extraction unit 602 extracts the blood vessel region 422 from the fundus photograph 421 and outputs it to the blood vessel alignment unit 603. The method for extracting the blood vessel region 422 in step S903 is the same as that in step S902. That is, by analyzing the pixel values of the fundus photograph 421, calculating the difference in the pixel values of the adjacent pixels, and searching for the adjacent pixels in which the difference is larger than the predetermined value, the blood vessel region and the other region can be separated from each other. Detect the boundary area. Thereby, the blood vessel region can be extracted from the fundus photograph 421. Then, in step S905, the blood vessel alignment portion 603 aligns the blood vessel region 412 and the blood vessel region 422.

In the following step S906, the blood vessel alignment unit 603 integrates the blood vessel region 412 of the fundus photograph 421 and the blood vessel region 422 of the integrated image 411 using the alignment result, and passes the integrated blood vessel region to step S803. The integration of the vascular region 412 and the vascular region 422 can be achieved, for example, by calculating the logical product or OR of the two regions. If the fundus photograph 421 is not input, the blood vessel region 412 extracted from the integrated image 411 is passed to step S803. This is the end of step S802.

As described above, in the present embodiment, the finding that the blood vessels of the fundus run near the lower end of the nerve fiber layer is expressed by the distance transfer function, and the feature that the brightness value of the fundus blood vessels is high in the retinal tomographic image is the brightness transfer function. Express with and volume render. As a result, the three-dimensional running of the fundus blood vessels of the retinal volume data can be visualized.

[Second Embodiment: Visualization of vitiligo] In the first embodiment described above, the blood vessel region and the layer boundary are extracted from the fundus image, and the blood vessel is volume-rendered using the opacity calculated from the distance from the layer boundary and the brightness value of the retinal volume data. Visualized the three-dimensional running of. On the other hand, in the second embodiment, it is an object to visualize vitiligo, which is a characteristic lesion of the fundus, instead of blood vessels.

The configuration of the fundus image display device and the device connected to the fundus image display device according to the present embodiment is the same as the configuration of the first embodiment shown in FIG. However, the functional block of the fundus image display device of the present embodiment is configured by adding the vitiligo region extraction unit 1001 to the functional configuration of the first embodiment as shown in FIG.

In the following, the white spot region extraction unit 1001 and the white spot center search unit 704, which is one of the functional configurations of the distance transfer function setting unit 505, will be described, and the other functional configurations will be described for the same processing as in the first embodiment. Omit.

The vitiligo region extraction unit 1001 extracts the vitiligo region from the input fundus image. When the input fundus image is retinal volume data 401, the vitiligo region is extracted based on the information of the retinal layer and the brightness value. Here, the position where vitiligo appears can be predicted based on the medical finding that it is near the outer plexiform layer 406. To utilize this finding, first, the nerve fiber layer 405 and the retinal pigment epithelium 407 are extracted. Furthermore, since the white spot in the retinal tomographic image 402 has a higher brightness value than the surroundings, the high brightness region between the nerve fiber layer 405 and the retinal pigment epithelium 407 is extracted by image processing such as binarization, and the retinal volume. Let it be the white spot area of data 401.

In addition, when the fundus photograph 421 is input in addition to the retinal volume data 401, the fundus photograph 421 may capture the white spots better. Therefore, in the fundus photograph 421, the two-dimensional white spot region is extracted first. To do. The extraction method here can also be realized by performing image processing such as binarizing the high-luminance region by utilizing the fact that vitiligo has a higher luminance value than the surroundings.

Next, based on the alignment information that can be calculated by the blood vessel region extraction unit 503, the white spot projection region similar to the blood vessel projection region is obtained by back-projecting the white spot region of the fundus photograph 421 onto the retinal volume data 401. Further, as described above, the vitiligo region is extracted from the voxels in the vitiligo projection region. By this method, it is possible to include the vitiligo that does not appear in the retinal volume data 401 but appears in the fundus photograph 421 in the vitiligo region.

The white spot center search unit 704 of the distance transfer function setting unit 505 searches for voxels in the white spot region in the z-axis direction in the white spot region extracted by the white spot region extraction unit 1001. At that time, one layer boundary is selected from the layer boundary information, the distance from the selected layer boundary (for example, the inner limiting membrane 404) to the voxel existing at the midpoint of the white spot region is calculated, and the peak position of the distance transfer function is calculated. To do.

Next, a procedure for visualizing vitiligo in the retinal volume data will be described using the flowchart shown in FIG. The process shown by the flowchart of FIG. 11 is realized by the CPU 5 executing the program stored in the main memory 6. Step S1101 and step S1102, step S1103, step S1105, and step S1107 are the same as step S801 and step S802, step S803, step S804, and step S806 in FIG. 8, respectively. Therefore, the description here will be omitted.

In step S1104, when the input fundus image is only the retinal volume data 401, the three-dimensional vitiligo region is extracted from the retinal volume data 401. When the fundus photograph 421 is input in addition to the retinal volume data 401, the vitiligo region is first extracted from the fundus photograph 421. Next, the vitiligo region of the fundus photograph 421 is back-projected onto the retinal volume data 401 based on the blood vessel region alignment information calculated by the blood vessel alignment portion 603. Further, the three-dimensional region of vitiligo is extracted using the layer information extracted in step S1103. The extracted white spot area information is output to the distance calculation unit 504.

In step S1106, the luminance transfer function setting unit 506 and the distance transfer function setting unit 505 set the luminance transfer function and the distance transfer function. The distance transfer function setting unit 505 obtains a predetermined position corresponding to the center of the white spot in the extracted white spot region. Then, the predetermined position of the z coordinate on each (x, y) coordinate of the vitiligo region from the inner limiting membrane 404 is set as the peak position of the distance transfer function. The set luminance transfer function and distance transfer function are output to the visualization unit 507.

As described above, in the second embodiment, the finding that white spots appear near the outer plexiform layer is expressed by the distance transfer function, and the feature that the brightness value of the white spots is high in the retinal tomographic image is expressed by the brightness transfer function. Volume rendering. This makes it possible to visualize the three-dimensional distribution of vitiligo in the retinal volume data.

[Third Embodiment] The configuration of the fundus image display device and the device connected to the fundus image display device according to the present embodiment is the same as the configuration of the first embodiment shown in FIG. Further, the basic configuration of the computer for realizing the functions of each part of the fundus image display device according to the present embodiment by software is the same as the configuration of the first embodiment shown in FIG.

As shown in FIG. 7, the distance transfer function setting unit 505 of the first embodiment and the second embodiment includes an average network thickness acquisition unit 701, a function parameter input unit 702, a shadow region extraction unit 703, and a white spot center search. A unit 704 and a layer boundary acquisition unit 705 are provided. In the first embodiment, the shadow region extraction unit 703 is used, and in the second embodiment, the vitiligo center search unit 704 is used to set the distance transfer function. In this embodiment, a distance transfer function setting method using any one of the average network thickness acquisition unit 701, the function parameter input unit 702, and the layer boundary acquisition unit 705 will be described.

The average net thickness acquisition unit 701 acquires the average thickness of each layer from the database 2 and determines a predetermined position corresponding to the peak position of the distance transfer function based on the average thickness. When visualizing the fundus blood vessels, the average thickness of the nerve fiber layer 405 is obtained because the fundus blood vessels appear near the lower end of the nerve fiber layer 405. When visualizing white spots, since white spots appear near the outer plexiform layer 406, the average thickness from the inner limiting membrane 404 to the upper end of the outer plexiform layer 406 is obtained. The predetermined position having the acquired thickness is set as the peak position of opacity, and the distance transfer function is set so that the opacity becomes 1.0.

The function parameter input unit 702 manually sets the peak position using the user interface. The layer boundary acquisition unit 705 acquires boundary information of each layer from the layer extraction unit 502 and sets it as the peak position of the distance transfer function. When visualizing the fundus blood vessels, the boundary information of the nerve fiber layer is acquired, and the peak position is set so that the thickness of the nerve fiber layer becomes opacity = 1.0. When visualizing vitiligo, the boundary information of the outer plexiform layer is acquired, and the peak position is set so that the thickness from the inner limiting membrane to the upper end of the outer plexiform layer has opacity = 1.0.

In this way, one of the average network thickness acquisition unit 701, the function parameter input unit 702, the shadow area extraction unit 703, the white spot center search unit 704, and the layer boundary acquisition unit 705, which constitute the distance transfer function setting unit 505, is used. Use to set the distance transfer function. This makes it possible to visualize the three-dimensional running of the fundus blood vessels and the three-dimensional distribution of vitiligo.

[Other Embodiments] Next, a basic configuration of a computer for realizing the functions of each block of the fundus image display device of the first embodiment, the second embodiment, and the third embodiment by a computer program will be described with reference to FIG.

The CPU 5 controls the entire computer using the data and programs stored in the main memory 6. In addition, the execution of software corresponding to each part of the fundus image display device 1 is controlled to realize the function of each part.

The main memory 6 stores the control program executed by the CPU 5 and provides a work area when the program is executed by the CPU 5.

The magnetic disk 7 stores an operating system (OS), a device drive of a peripheral device, various application software including a program for displaying a fundus image, and the like. The display memory 8 temporarily stores display data for the monitor 9.

The monitor 9 is, for example, a CRT monitor, a liquid crystal monitor, or the like, and displays an image based on the data from the display memory 8. The mouse 10 and the keyboard 11 each perform pointing input and character input by the user. The operator can use these to give various instructions to the fundus image display device 1. Each of the above components is communicatively connected to each other by a common bus 12.

Although the embodiments have been described in detail above, the present invention can be implemented as, for example, a system, an apparatus, a method, a program, a storage medium, or the like. Specifically, it may be applied to a system composed of a plurality of devices, or may be applied to a device composed of one device.

In the present invention, the function of the above-described embodiment is achieved by supplying a software program to the system or device directly or remotely, and the computer of the system or device reads and executes the supplied program code. including. In this case, the supplied program is a computer program corresponding to the flowchart shown in the figure in the embodiment.

Therefore, in order to realize the functional processing of the present invention on a computer, the program code itself installed on the computer also realizes the present invention. That is, the present invention also includes a computer program itself for realizing the functional processing of the present invention. In that case, as long as it has a program function, it may be in the form of object code, a program executed by an interpreter, script data supplied to the OS, or the like.

Computer-readable storage media for supplying computer programs include: For example, floppy (registered trademark) disks, hard disks, optical disks, optical magnetic disks, MOs, CD-ROMs, CD-Rs, CD-RWs, magnetic tapes, non-volatile memory cards, ROMs, DVDs (DVD-ROMs, DVD-)s. R) and so on.

In addition, the program supply method includes connecting to a home page of the Internet using a browser of a client computer and downloading the computer program of the present invention to a recording medium such as a hard disk from the home page. In this case, the downloaded program may be a compressed file containing an automatic installation function. It can also be realized by dividing the program code constituting the program of the present invention into a plurality of files and downloading each file from different homepages. That is, the present invention also includes a WWW server that allows a plurality of users to download a program file for realizing the functional processing of the present invention on a computer.

Further, the program of the present invention may be encrypted, stored in a storage medium such as a CD-ROM, and distributed to users. In this case, let the user who clears the predetermined conditions download the key information to decrypt from the homepage via the Internet, execute the encrypted program using the key information, and install the program on the computer. Can also be done.

In addition, when the computer executes the read program, the functions of the above-described embodiment are realized, and based on the instructions of the program, the embodiment is collaborated with the OS running on the computer. The function may be realized. In this case, the OS or the like performs a part or all of the actual processing, and the processing realizes the functions of the above-described embodiment.

Further, the program read from the recording medium is written to the memory provided in the function expansion board inserted in the computer or the function expansion unit connected to the computer, and some or all of the functions of the above-described embodiment are realized. You may. In this case, after the program is written to the function expansion board or the function expansion unit, the CPU provided in the function expansion board or the function expansion unit performs a part or all of the actual processing based on the instruction of the program.

<figref num="1">It is a figure which shows an example of the transfer function for volume rendering.</figref><figref num="2">It is a figure for demonstrating a retinal tomographic image.</figref><figref num="3">It is a figure which shows an example of the structure of the fundus image display system which concerns on embodiment of the invention.</figref><figref num="4">It is a figure for demonstrating the relationship between the fundus photograph, the retinal volume data and the integrated image which concerns on embodiment of an invention.</figref><figref num="5">It is a figure which shows an example of the functional structure of the fundus image display device 1 corresponding to the embodiment of the invention.</figref><figref num="6">It is a figure which shows an example of the functional structure of the blood vessel region extraction part 503 corresponding to the embodiment of the invention.</figref><figref num="7">It is a figure which shows an example of the functional structure of the distance transfer function setting unit 505 corresponding to the embodiment of the invention.</figref><figref num="8">It is a flowchart which shows the processing procedure which visualizes the 3D running of the fundus blood vessel in the fundus image display apparatus 1 of 1st Embodiment of this invention by volume rendering.</figref><figref num="9">It is a flowchart which shows the detail of the extraction process of the blood vessel region by the blood vessel region extraction unit 503 in the 1st Embodiment of the invention.</figref><figref num="10">It is a figure which shows an example of the functional structure of the fundus image display device 1000 which concerns on 2nd Embodiment of this invention.</figref><figref num="11">It is a flowchart which shows the processing procedure which visualizes the vitiligo of the retinal volume data in the fundus image display apparatus 1000 which concerns on 2nd Embodiment of this invention.</figref><figref num="12">It is a figure which shows an example of the hardware composition of the fundus image display device which concerns on embodiment of an invention.</figref>

Code description

1 Fundus image display device 2 database 3 LAN 4 Fundus image display device 5 CPU 6 Main memory 7 magnetic disk 8 Display memory 9 monitor 10 mouse 11 keyboard 12 common bus