JP2011212242A - Virtual endoscope apparatus with entire sphere type visual field - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a virtual endoscope apparatus forming a virtual endoscope image with a maximum view angle of 90° or more.SOLUTION: An entire sphere type virtual endoscope image having a visual field that surrounds a visual point position spherically is formed by an entire sphere type image forming means 58 of a virtual endoscope image forming means 56, and the entire sphere type virtual endoscope image is displayed on an output device 26 by a display means 64 to obtain the entire sphere type endoscope image having a wide visual field that surrounds the visual point position spherically by the virtual endoscope apparatus 10.

Description

  The present invention relates to a virtual endoscope apparatus that virtually generates an endoscopic image of a subject based on three-dimensional image data of the subject, and in particular, a global type having a visual field surrounding a viewpoint position in a spherical shape. The present invention relates to a virtual endoscope apparatus capable of generating a virtual endoscopic image.

  In recent years, in the medical field, diagnosis and examination using an image of a subject obtained by an X-ray CT (computed tomography) apparatus, a magnetic resonance imaging (MRI) apparatus, or the like has been widely performed. In the CT apparatus, by continuously feeding the subject in the body axis direction while continuously rotating the X-ray irradiator / detector, a spiral continuous scan (helical scan) is performed on the three-dimensional region of the subject. Is done. The grayscale image data (three-dimensional image data) in the three-dimensional region thus obtained is, for example, an opacity that gives a spatial distribution state of density values of the three-dimensional grayscale image to each sample point of the three-dimensional image. The image is visualized as a three-dimensional image by a technique called volume rendering, which is a method of drawing from color information.

  According to the three-dimensional image, it is possible to virtually generate an endoscope image of the subject based on the three-dimensional image data obtained by the X-ray CT apparatus. A virtual endoscope apparatus capable of generating such a virtual endoscopic image has been proposed. For example, this is the virtual endoscope apparatus described in Non-Patent Document 1. According to such a virtual endoscope apparatus, since an endoscopic image can be obtained without actually inserting an endoscope into the subject, the burden on the subject is reduced. On the other hand, a doctor or the like who performs an examination can obtain an image similar to that of an endoscope, and thus can perform an examination or the like with the same senses and skills as those conventionally performed using an endoscope.

Kensaku Mori, Akihiro Urano, Jun-ichi Hasegawa, Jun-ichiro Toriwaki, Hirofumi Anno and Kazuhiro Katada, “Virtualized circular system-an application of virtual reality technology to diagnostic aid,” IEICE Transactions Information and Systems, June 1996. Vol.E79-D, No.6, p.809-819 Masahiro Oda, Takayuki Kitasaka, Yuichiro Hayashi, Kensaku Mori, Yasuhito Suenaga, and Jun-ichiro Toriwaki, “Development of a Navigation-Based CAD System for Colon,”, 8th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2005) , Palm Springs, CA, USA, October 26-30, 2005, Proceedings Part I, LNCS 3749, Gerhard Goos, Juris Hartmanis, and Jan van Leeuwen (Eds.), October 2005, pp.696-703 Kensaku Mori, Yasuhito Suenaga, Junichiro Toriwaki, Makoto Hashizume, “Study on virtual laparoscopic image creation for laparoscopic surgery support,” Japan Society for Computer Aided Surgery, September 2005, Vol. 7, No. 2, pp.95-104

  By the way, in many conventional virtual endoscope apparatuses, a fluoroscopic method (perspective projection method) having a maximum viewing angle of, for example, about 100 degrees to 130 degrees is used. In this way, a virtual endoscopic image approximate to an endoscopic image optically obtained using an actual endoscope can be obtained. On the other hand, as described in Non-Patent Document 2, when the virtual endoscope apparatus is applied to a colon polyp diagnosis system, for example, there are many folds in the large intestine, which is one part of the subject, When viewed from the viewpoint, there is a possibility that a lesion existing in a position hidden behind the folds may be overlooked. In addition, as described in Non-Patent Document 3, when the virtual endoscope apparatus is used as a virtual laparoscope, in the operation support such as determination of the insertion position of a forceps or the like, a virtual view with a narrow viewing angle can be obtained. The endoscopic image was insufficient.

  However, if the viewing angle is increased in order to widen the range that can be observed at one time, the distortion in the virtual endoscopic image becomes very large, and the virtual endoscopic image cannot be used for examination or surgery support. There was a problem of becoming something. Further, according to the fluoroscopic method, it was theoretically impossible to set the maximum viewing angle to 180 degrees or more.

  The present invention has been made against the background of the above circumstances, and an object of the present invention is to provide a virtual endoscope apparatus capable of generating a virtual endoscope image having a maximum viewing angle of 180 degrees or more. is there.

  In order to achieve this object, the invention according to claim 1 is based on the three-dimensional image data of the subject, and virtually stores a virtual endoscopic image obtained when the endoscope is inserted into the subject. A virtual endoscopic apparatus comprising: a virtual endoscopic image generating means for generating; and a display means for displaying at least a virtual endoscopic image generated by the virtual endoscopic image generating means, wherein the virtual endoscope The mirror image generation means generates a global virtual endoscopic image having a field of view surrounding the viewpoint position in a spherical shape.

  According to a second aspect of the present invention, in the virtual endoscope device according to the first aspect, the virtual endoscopic image generation means generates the ray from the viewpoint position to the field of view and uses the volume rendering method to generate the global Generating a type virtual endoscopic image.

  According to a third aspect of the present invention, in the virtual endoscopic device according to the first or second aspect, the virtual endoscopic image generating means generates a global virtual endoscopic image having a preset maximum viewing angle. It is characterized by.

  According to a fourth aspect of the present invention, in the virtual endoscopic device according to any one of the first to third aspects, the virtual endoscopic image generating means expands or compresses a specific angular region at the viewing angle. A global virtual endoscopic image is generated.

  The invention according to claim 5 is the virtual endoscope apparatus according to claim 4, characterized in that the position and range of the specific angle region can be changed.

  According to a sixth aspect of the present invention, in the virtual endoscope device according to the fourth or fifth aspect, the characteristic part extracting means for extracting a characteristic part including a preset characteristic from the three-dimensional image data of the subject, A specific region setting means for setting the specific angle region so as to include the feature portion when the feature portion extracted by the feature portion detection means is included in the global virtual endoscopic image; The virtual endoscopic image generation means enlarges and displays the specific angle area set by the specific area setting means.

  According to a seventh aspect of the present invention, in the virtual endoscope apparatus according to any one of the first to sixth aspects, the virtual endoscopic image generating means corresponds to the fact that the viewpoint position is continuously changed. A global virtual endoscopic image can be continuously generated.

  According to an eighth aspect of the present invention, in the virtual endoscope device according to any one of the first to seventh aspects, the display means superimposes the global virtual endoscopic image and a symbol for displaying a viewing angle. Display.

  According to a ninth aspect of the present invention, in the virtual endoscope device according to any one of the first to eighth aspects, the virtual endoscopic image generation means is configured to add the global virtual endoscopic image from the viewpoint position. A fluoroscopic virtual endoscopic image generated by a fluoroscopic method can be generated, and the display means switches between the global virtual endoscopic image and the fluoroscopic virtual endoscopic image for display; It is characterized by.

  According to a tenth aspect of the present invention, the virtual endoscopic image generating means can generate a fluoroscopic virtual endoscopic image generated by a fluoroscopy method from the viewpoint position in addition to the global virtual endoscopic image. And the display means displays the global virtual endoscopic image and the fluoroscopic virtual endoscopic image in a comparable manner.

  According to the first aspect of the present invention, a global virtual endoscopic image having a field of view surrounding the viewpoint position in a spherical shape is generated by the virtual endoscopic image generating means, and the global virtual endoscopic image is generated by the display means. Since a mirror image is displayed, a global virtual endoscopic image having a wide field of view that surrounds the viewpoint position in a spherical shape can be obtained by the virtual endoscopic device.

  According to the invention of claim 2, the virtual endoscopic image generating means generates the global virtual endoscopic image using a volume rendering method by generating a ray from the viewpoint position to the visual field. Therefore, a global virtual endoscopic image having a wide field of view that surrounds the viewpoint position in a spherical shape can be generated based on the three-dimensional image data.

  According to the invention of claim 3, the virtual endoscopic image generating means generates a global virtual endoscopic image with a preset viewing angle, so that a field of view surrounding the viewpoint position in a spherical shape is set in advance. A global virtual endoscopic image with a wide viewing angle can be obtained.

  According to the invention of claim 4, the virtual endoscopic image generation means generates the global virtual endoscopic image obtained by expanding or compressing a specific angular region at the viewing angle, so the viewpoint position is determined. It is possible to obtain the global virtual endoscopic image that has a wide field of view surrounding a sphere and is enlarged or reduced with respect to an angle range of a specific viewing angle.

  According to the fifth aspect of the present invention, the position and range of the specific angle area can be changed. For example, the global area in which the angle area is enlarged by designating the specific angle area that the operator wants to enlarge and display. Type virtual endoscopic images can be obtained.

  According to the sixth aspect of the present invention, a feature portion including a preset feature is extracted from the three-dimensional image data of the subject by the feature portion extraction means, and the feature portion is extracted into the global virtual endoscopic image. If the specified feature region is included, the specific region setting unit sets the specific angle region so as to include the feature region, and the virtual endoscope image generation unit sets the specific region by the specific region setting unit. Since the specific angle area is enlarged and displayed, the global virtual endoscope has a wide field of view that surrounds the viewpoint position in a spherical shape, and a feature portion including a preset feature is enlarged and displayed. A mirror image can be obtained.

  According to the seventh aspect of the invention, since the virtual endoscopic image generation means continuously generates the global virtual endoscopic image in response to continuously changing the viewpoint position. The global virtual endoscopic image corresponding to the continuously changing viewpoint position and having a wide field of view surrounding the viewpoint position in a spherical shape can be obtained.

  According to the invention of claim 8, the display means displays the global virtual endoscopic image and the symbol for displaying the viewing angle so as to overlap each other, so that a wide field of view that surrounds the viewpoint position in a spherical shape is displayed. The viewing angle can be grasped in the global virtual endoscopic image.

  According to the invention according to claim 9, in addition to the global virtual endoscopic image, the virtual endoscopic image generation unit generates a fluoroscopic virtual endoscopic image generated by the fluoroscopy from the viewpoint position. Since the global virtual endoscopic image and the perspective virtual endoscopic image are switched and displayed by the display means, the display means has a wide field of view that surrounds the viewpoint position in a spherical shape at the same viewpoint position. It is possible to switch between a global virtual endoscopic image and the fluoroscopic virtual endoscopic image close to an actual endoscopic image for observation.

  According to the tenth aspect of the present invention, in addition to the global virtual endoscopic image, the virtual endoscopic image generation unit generates a fluoroscopic virtual endoscopic image generated by the fluoroscopy from the viewpoint position. Since the global virtual endoscopic image and the fluoroscopic virtual endoscopic image are generated and displayed by the display means in a comparable manner, the virtual endoscopic image generated by a different method from a common viewpoint position is generated. The mirror image can be referred to at the same time, and convenience in observation of the virtual endoscopic image is improved.

  Preferably, the viewing angle of the global virtual endoscope can be continuously changed. In this way, virtual endoscopic images whose viewing angles continuously change at the same viewpoint position are sequentially obtained.

  Preferably, the display means switches between a global virtual endoscopic image and a perspective virtual endoscopic image according to the viewing angle and displays the switched virtual endoscopic image. In this way, the virtual endoscopic image corresponding to the viewing angle can be automatically switched and displayed.

  Preferably, the display means switches from a fluoroscopic virtual endoscopic image to a global virtual endoscopic image when a characteristic part is extracted, and displays it. In this way, when the feature portion is displayed, the feature portion is switched to the global virtual endoscopic image that can be enlarged and displayed.

  Further preferably, when the viewing angle of the generated global virtual endoscopic image is equal to or larger than a preset boundary angle, the display means has a viewing angle for a region where the viewing angle is equal to or larger than the boundary angle. A global virtual endoscopic image is displayed with a hue different from that of an area less than the boundary angle. In this way, the hue of the region whose viewing angle is equal to or larger than the boundary angle, that is, the region corresponding to the rear of the viewpoint position in the global virtual endoscopic image, is the region whose viewing angle is less than the boundary angle, that is, the global virtual Since it is displayed differently from the hue of the region corresponding to the front of the viewpoint position in the endoscopic image, it is viewed in an actual endoscopic image or a perspective type virtual endoscope that has been widely used in the past. A region corresponding to the rear of the viewpoint position that could not be identified can be distinguished in the global virtual endoscopic image.

It is a figure explaining the structure of the virtual endoscope apparatus which is one Example of this invention. It is a block diagram explaining the outline | summary of the function which the virtual endoscope apparatus of FIG. 1 has. It is a figure explaining the concept of a global virtual endoscopic image. It is a figure explaining the production | generation of the global virtual endoscopic image whose maximum viewing angle (PHI) is 360 degree | times. It is a figure explaining the concept of a perspective type virtual endoscopic image. It is a flowchart explaining the outline | summary of the control action of the global type | mold virtual endoscope image by the virtual endoscope apparatus in an Example. It is a figure explaining the definition of a projection surface and each coordinate. It is a flowchart explaining the outline | summary of the control action in a ray direction calculation subroutine. It is a flowchart explaining the outline | summary of the control action in (theta) 1 calculation subroutine. It is a flowchart explaining the outline | summary of the control action | operation in (theta) 2 calculation subroutine. It is a flowchart explaining the outline | summary of the control action | operation in a shadow value calculation subroutine. It is a figure explaining the example of a normalization coordinate system. It is a figure explaining an example of the function f (l) showing the relationship of rotation angle (theta) 1 with respect to the distance l of the coordinate position in a normalization coordinate system, and an origin. It is a figure explaining the overlap of a ray and a three-dimensional image. It is a figure explaining the relationship between the coordinate of a sample point and eight points surrounding it, Comprising: It is a figure explaining calculation of opacity, or calculation of a shadow value. It is a figure explaining the example by which a global virtual endoscopic image and the display showing a viewing angle are superimposed and displayed. It is the figure which showed the virtual laparoscopic image which is an example of the virtual endoscopic image obtained by the virtual endoscopic apparatus of this invention about several viewing angles. It is the figure which showed the colon endoscope image which is an example of the virtual endoscopic image obtained by the virtual endoscopic apparatus of this invention about several viewing angles. It is a figure explaining the case where a perspective type virtual endoscopy image and a global type virtual endoscopy image are displayed by contrast display means so that contrast is possible. It is a flowchart explaining an example of the control action | operation in the display of the symbol showing the viewing angle by a viewing angle display means. It is a flowchart explaining an example of the control action in a display switching means. It is a figure explaining the example of the normalization coordinate system in another Example of this invention, Comprising: It is a figure corresponding to FIG. It is a figure explaining the example of the function f (l) in another Example of this invention, Comprising: It is a figure corresponding to FIG. It is a figure explaining the example which sets the function f (l) arbitrarily in another Example of this invention. It is a figure which shows an example of the virtual endoscopic image from which the hue of the area | region exceeding the magnitude | size of the said boundary angle and the hue of the area | region less than the boundary angle differed by the viewing angle display means.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

  As shown in FIG. 1, the virtual endoscope apparatus 10 of the present embodiment includes a ROM 14 that stores information stored in a readable manner, a RAM 16 that stores information in a readable and writable manner as necessary, and stores these information. It is constituted by a CPU (central processing unit) 18 or the like that performs computation, and is constituted by a so-called computer 12 that can perform computation control according to a program stored in advance. In the computer 12, an input device 24 such as a keyboard and a mouse is connected via the input interface 20, and an input operation on the computer 12 can be performed. Further, an output device 26 such as a display device connected via the output interface 22 is connected, and the output of the computer 12 can be output visually. For example, a storage device 28 such as a hard disk drive capable of storing information is provided therein. The storage device 28 stores, for example, a three-dimensional gray image f of a subject obtained by a CT device, in other words, a pixel value (density value) for each voxel that is a unit volume of the subject.

  FIG. 2 is a block diagram for explaining the outline of the control function in the virtual endoscope apparatus 10, that is, the outline of the control function obtained by executing a predetermined program in the computer 12 including the CPU 18. Among these, the viewpoint position setting means 50 sets the viewpoint position p and the line-of-sight direction L of the virtual endoscopic image obtained by the virtual endoscope apparatus 10 based on an operation using the operator's input device 24. This line-of-sight direction L is defined as the center of the visual field, for example. In addition, the viewpoint position setting unit 50 can change the viewpoint position p and the line-of-sight direction L so as to dynamically change based on an operation performed via the input device 24.

  The visual field setting unit 52 sets the maximum visual field angle Φ of the virtual endoscopic image obtained by the virtual endoscope device 10 based on an operation using the operator's input device 24. The maximum viewing angle Φ corresponds to Φ in FIG. 3 or FIG. 5 described later, and in the virtual endoscopic image, a pair of directions farthest from the line-of-sight direction L determined by the viewpoint position setting unit 50 is present. It is an angle to make. Therefore, the viewing angle in the virtual endoscopic image corresponds to an angle twice as large as the angle formed between the viewing direction L and the direction farthest from the viewing direction in the virtual endoscopic image. The viewing angle setting means 52 can set the maximum viewing angle Φ so as to dynamically change based on an operation performed via the input device 24. In the global virtual endoscopic image generated by the global image generating means 58 of the virtual endoscopic image generating means 56 described later, the size of the maximum viewing angle Φ is not particularly limited. On the other hand, in a fluoroscopic endoscope image generated by the fluoroscopic image generating means 62, generally, when the maximum viewing angle Φ exceeds 90 degrees, image distortion increases, and as the maximum viewing angle Φ approaches 180 degrees. The image diverges. Also, it is theoretically impossible for the maximum viewing angle Φ to be 180 degrees or more.

  The specific area setting unit 54 sets a specific angle range as the specific area SR among the viewing angles in the virtual endoscopic image. Specifically, for example, when the viewing angle is 179 degrees, a specific angle range representing the specific region SR is set such that 180 degrees ≦ θ ≦ 220 degrees. θ is an angle from the viewing direction. The setting of the specific area SR by the specific area setting means 54 is performed based on an operation using the input device 24 by the operator. Or it can also set automatically so that the feature part extraction means 72 mentioned later may be included. Specifically, for example, the specific region SR is set so as to include a feature part extracted by a feature part extraction unit 72 described later and a predetermined margin. More specifically, when the feature part is extracted in the range of 188 degrees to 192 degrees in the virtual endoscopic image, if the predetermined margin is determined to be 4 degrees, the feature part extraction means 72 Sets the specific region SR using the θ as in an angle range of 184 degrees ≦ θ ≦ 196 degrees. The predetermined margin is a value determined so that the feature part is displayed so as to be recognized well in the virtual endoscopic image, and is determined in advance as described above. The value may be fixed, or may be variable as the subsequent generation of the virtual endoscopic image is performed. The specific area SR set by the specific area setting means 54 corresponds to a specific angle range of the present invention.

  The virtual endoscopic image generation unit 56 is set by the three-dimensional image data of the subject stored in the storage unit 28, the viewpoint position and the line-of-sight direction set by the viewpoint position setting unit 72, and the visual field setting unit 52. Based on the maximum viewing angle Φ, a virtual endoscopic image of the subject is generated. That is, the virtual endoscopic image generation means 56 generates an endoscopic image obtained when an endoscope is inserted so as to be in the visual line direction at the viewpoint position of the subject based on the three-dimensional image data. To generate virtually. In the present embodiment, the virtual endoscope image generating means 56 includes a global image generating means 58 that generates a global virtual endoscopic image and a fluoroscopic image that generates a perspective virtual endoscopic image using a perspective method. The mold image generating means 62 is functionally included.

  Among these, the global image generation means 58 generates a global virtual endoscopic image by the global projection method. The global virtual endoscopic image is a virtual endoscopic image having a field of view that surrounds the viewpoint position set by the viewpoint position setting means 72 in a spherical shape. The maximum viewing angle Φ in this global virtual endoscopic image is set by the visual field setting means 52.

  FIG. 3 is a diagram illustrating an outline of generation of a global virtual endoscopic image, and is a diagram illustrating a relationship between a field of view in one plane including the viewpoint position p and pixels on the projection plane. When the viewpoint position p and the line-of-sight direction L are set as shown in FIG. 3, a virtual projection plane 102 that surrounds the viewpoint position p in a spherical shape with the viewpoint position p as the center is virtually set. At this time, the line-of-sight direction L is set to face the center of the projection plane 102. Then, a ray is directed from the viewpoint position p to each direction of a predetermined number of pixels arranged on the virtual projection plane 102, and the shadow value of each sample point on the ray is multiplied by opacity. A so-called ray casting method for calculating a pixel value in each of the pixels is performed by accumulating. This ray casting method is one method of a so-called volume rendering method.

  When the execution of the ray casting method is completed for one plane including the viewpoint position p as shown in FIG. 3, the global image generation unit 58 sets the plane including the viewpoint position p at a predetermined angle about the line-of-sight direction L. Rotate. Then, the ray casting method is similarly executed on the rotated plane. By repeating this until the plane is rotated by π around the line-of-sight direction L, a global virtual endoscopic image is obtained. The predetermined angle is appropriately determined according to the number of pixels on the projection surface, for example.

  FIG. 4 is another diagram for explaining the generation of a global virtual endoscopic image when the maximum viewing angle Φ is 360 degrees. FIG. 4A is a diagram for explaining the rotation of the plane around the line-of-sight direction L, and FIG. 4B shows the rotation plane and the pixels on the global virtual endoscopic image. It is a figure explaining a relationship. In FIG. 4A, three orthogonal directions V, U, and R are set. The directions V, U, and R are set to be the same as, for example, directions of a visual axis direction vector d_v, an upward direction vector d_u, and a rightward direction vector d_r described later (see FIG. 7).

  In FIG. 4, three F1 to F3 are described as an example of a plane including the viewpoint position p corresponding to FIG. An angle θ1 representing the direction of the ray in the plane Fn (n = 1, 2,...) Is defined as an angle formed by the ray and the V direction that is the line-of-sight direction L. That is, when the ray and the line-of-sight direction L coincide with each other, θ1 = 0 degree, and when viewed in the direction opposite to the U direction, that is, when viewed from above in FIG. Is defined. Further, as shown in FIG. 4, the rotation angle θ2 around the viewing direction of each plane Fn, that is, around the V direction, is a plane Fn, a plane including the V direction and the R direction (hereinafter referred to as “VR plane”). Is defined as the angle formed by. That is, the case where the plane Fn is in the VR plane is defined as θ2 = 0 degrees, and the direction rotating counterclockwise around the V direction is defined as positive.

  As indicated by a bold line in FIG. 4A, the direction of the ray can be expressed using θ1 and θ2 defined as described above. In FIG. 4A, when the plane Fn including the viewpoint position p is determined by being rotated by a predetermined angle in the positive direction of θ2, θ1 is set to a predetermined value between −π and π in the plane, for example. Ray directions are sequentially determined at intervals, and pixel values are calculated for the directions by the ray casting method. Then, the plane is again rotated by a predetermined angle in the positive direction of θ2, and similarly, pixel values are calculated for each of the ray directions in the plane. By repeating this from θ2 to 0 ° to 360 °, a field of view that surrounds the viewpoint position p in a spherical shape, that is, a pixel value corresponding to a maximum viewing angle Φ of 360 ° is obtained. That is, when the plane is F1 (when θ2 = 0 degrees), the ray direction is set to each of the directions from the viewpoint position p toward each point represented by a triangle on the plane F1. Further, when the plane is F2, the direction from the viewpoint position p to each point represented by a circle on the plane F2 is represented by a rectangle on the plane F3 from the viewpoint position p when the plane is F3. Similarly, the direction of the ray is set for each of the directions toward each point.

  This procedure is performed after the ray casting method is executed so that θ2 is changed from 0 degree to 360 degrees with respect to one of the ray directions θ1 set at a predetermined interval between −π and π. , Θ1 may be updated and θ2 may be changed from 0 degree to 360 degrees again. In other words, the order of the pixel values is not limited as long as the pixel values can be calculated when the ray direction is directed at a predetermined interval in the field of view with the maximum viewing angle Φ set from the viewpoint position p.

  As shown in FIG. 4A, a circle with an appropriate radius centered on the viewpoint position p is set in the plane Fn. This circle corresponds to the virtual projection plane 102 in FIG. Further, when the plane Fn is sequentially rotated in the V direction, a sphere is formed by the rotation of the circle. This sphere corresponds to a visual field that surrounds the viewpoint position in the present invention in a spherical shape. Of the field of view that surrounds this viewpoint position in a spherical shape, a region within the maximum viewing angle Φ centered on the line-of-sight direction L is converted to the projection plane 104 as a global virtual endoscopic image.

  When the calculation of the pixel value for each pixel on the virtual projection plane 102 is completed, the global virtual endoscope image generation means 58 converts the virtual projection plane 102 into a projection plane 104 provided in a planar shape by a predetermined method. A global virtual endoscopic image is obtained. At this time, each pixel on the virtual projection plane is converted into each pixel on the planar projection plane 104 according to a predetermined relationship or the like. In this conversion, distortion on the projection plane 104 is reduced. Conversion is performed.

  FIG. 4B shows the projection surface 104 of the global virtual endoscopic image, and the relationship between the ray direction in FIG. 4A and the pixel position on the global virtual endoscopic image is shown. It is a figure explaining an example. The ray directions represented by circles, triangles, and squares in FIG. 4A correspond to the pixel positions represented by circles, triangles, and squares in FIG. 4B, respectively. ing. Specifically, each pixel on a straight line Qn (n = 1, 2,...) Passing through the origin on the projection surface 104 of the global virtual endoscopic image is on each plane Fn in FIG. It corresponds to each of Ray. At this time, the rotation angle θ2 around the origin from the axial direction of the straight line Qn passing through the origin on the projection plane 104 in FIG. 4B corresponds to the rotation angle θ2 around the visual line in FIG. 4A, that is, around the V direction. ing. Also, each pixel on the straight line Qn passing through the origin on the projection plane 104 in FIG. 4B has a distance between each pixel and the origin of the projection plane in the direction of each ray on the plane Fn in FIG. A predetermined relationship with θ1 is satisfied. This predetermined relationship is a relationship shown in FIG. 13, FIG. 23, or FIG. 24 described later. Specifically, when θ1 is 90 degrees, this corresponds to the coordinate origin on the projection plane 104. When θ1 is negative, they correspond to the negative region in the x-axis direction, and when θ1 is positive, they correspond to the positive region in the x-axis direction according to the predetermined relationship.

  Returning to FIG. 2, when the specific area SR, that is, the specific angle range is set by the specific area setting means 54, the specific area processing unit 60 expands (extends) the specific area SR in the global virtual image. The global image generation means 58 is controlled so as to be reduced (compressed) and displayed. Specifically, for example, when the specific region SR is enlarged (expanded) in the global virtual image, the global image generation unit 58 converts each pixel on the virtual projection surface 102 into a planar projection surface as described above. In the conversion to the 104 pixels, the conversion is performed so that the interval between the pixels within the specific angle range corresponding to the specific region SR is wider than the interval between the pixels outside the specific angle range. . More specifically, when converting each pixel on the virtual projection plane 102 to each pixel on the planar projection plane 104, the interval between the pixels within the specific angle range is out of the specific angle range. The interval between the pixels is increased compared with the interval between the pixels, the interval between the pixels outside the specific angle range is decreased as compared with the interval between the pixels within the specific angle range, or both. When the specific area SR is reduced (compressed) in the global virtual image, the global image generation unit 58 converts each pixel on the virtual projection plane 102 to each pixel on the planar projection plane 104 as described above. In the conversion, the conversion is performed so that the interval between the pixels within the specific angle range corresponding to the specific region is narrower than the interval between the pixels outside the specific angle range. More specifically, when converting each pixel on the virtual projection plane 102 to each pixel on the planar projection plane 104, the interval between the pixels within the specific angle range is out of the specific angle range. The interval between the pixels is narrowed compared with the interval between the pixels, the interval between the pixels outside the specific angle range is increased compared with the interval between the pixels within the specific angle range, or both.

  The fluoroscopic image generating means 62 generates a fluoroscopic virtual endoscope image. The perspective virtual endoscopic image is a virtual endoscopic image obtained by a perspective projection method when the maximum viewing angle Φ set by the visual field setting unit 52 is set from the viewpoint position p set by the visual point position setting unit 72. Corresponding to FIG. 5 is a diagram illustrating an outline of a perspective virtual endoscopic image. When the viewpoint position p and the line-of-sight direction L are set as shown in FIG. 5, the projection plane 106 is placed on a plane orthogonal to the line-of-sight direction L at a position away from the viewpoint position p in the line-of-sight direction L by a predetermined distance. Is provided. The size of the projection surface 106 is determined by the maximum viewing angle Φ. Then, a ray is directed to each direction of a predetermined number of pixels arranged on the projection plane 106 from the viewpoint position p, and the shadow value of each sample point on the ray is integrated as opacity. Thus, a so-called ray casting method for calculating a pixel value in each of the pixels is executed. When the calculation of the pixel value for each pixel on the projection plane 106 is completed in this way, a perspective virtual endoscopic image on the projection plane 106 is obtained. Specifically, the ray direction is always a straight line connecting the viewpoint position p and the pixel of interest on the projection plane 106 as compared to the generation of the global virtual endoscopic image by the global image generation means 58 (that is, This is different in that step S4 does not exist in the flowchart of FIG. In this perspective virtual endoscopic image, the interval between the pixels on the projection surface 106 with respect to a certain angle of the field of view becomes wider at the edge of the image as the maximum viewing angle Φ becomes wider, and thus image distortion is remarkable. It will be something.

  The display means 64 outputs the virtual endoscopic image generated by the virtual endoscopic image generation means 56 by displaying it on the output device 26. The display means 64 functionally includes a viewing angle display means 66, a display switching means 68, and a contrast display means 70, and the global type generated by the global image generation means 58 of the virtual endoscope image generation means 56. The virtual endoscope image or the perspective virtual endoscope image generated by the perspective image generation means 62 of the virtual endoscope image generation draft 56 is displayed as it is, and the display mode is displayed by each of these means. Can be switched.

  When the virtual endoscopic image generated by the virtual endoscopic image generating unit 56 is displayed on the output device 26, the viewing angle display unit 66 displays a symbol for displaying the viewing angle in the image. It is displayed superimposed on the mirror image.

  FIG. 16 shows an example of a global virtual endoscopic image generated by the global image generating means 58 of the virtual endoscopic image generating means 56, and further the global virtual endoscope It is also an example when the image and the symbol for displaying the viewing angle displayed by the viewing angle display means 66 are superimposed. This image is displayed on the output device 26. Since the maximum viewing angle Φ is 360 degrees in the global virtual endoscopic image of FIG. 16, in the global virtual endoscopic image represented in a rectangular shape in FIG. The angle corresponds to 360 degrees. In the example of FIG. 16, two viewing angles are displayed on the global endoscope image: a circle connecting points with a viewing angle of 180 degrees and a circle connecting points with a viewing angle of 320 degrees. It is displayed as a symbol. In addition, in order to identify these two circles, the angle between the direction corresponding to the circles and the visual axis direction, that is, the half of the viewing angle is indicated by a number in the vicinity of the circles. It is displayed.

  The display switching means 68 includes a global virtual endoscope image generated by the global image generation means 58 of the virtual endoscope image generation means 56 and a perspective virtual endoscope generated by the perspective image generation means 62. Switch between images. This switching display may be performed based on, for example, a user instruction from the input device 24, or automatically performed based on the maximum viewing angle Φ of the virtual endoscopic image, as will be described later. May be.

  FIG. 17 shows a virtual endoscopic image obtained when an image of a laparoscope inserted into the abdomen that has been ablated (inflated with air or the like) is generated by the virtual endoscopic device of the present invention. FIG. 17A is a perspective virtual endoscope image when the maximum viewing angle Φ is 130 degrees, and FIG. 17B is a perspective virtual endoscope when the maximum viewing angle Φ is 179 degrees. Endoscopic image, FIG. 17C shows a global virtual endoscopic image when the maximum viewing angle Φ is 360 degrees. FIG. 18 is a diagram showing a virtual endoscopic image obtained when an endoscope image inserted into the large intestine of a subject is generated by the virtual endoscopic device of the present invention. FIG. 18A is a perspective virtual endoscope image when the maximum viewing angle Φ is 112 degrees, and FIG. 18B is a perspective virtual endoscope image when the maximum viewing angle Φ is 179 degrees. c) shows a global virtual endoscopic image when the maximum viewing angle Φ is 360 degrees.

  As shown in FIGS. 17 (a) and 18 (a), when the maximum viewing angle Φ is less than 180 degrees, a fluoroscopic virtual endoscope image can be theoretically generated. The closer to the edge of the screen, the greater the distortion of the image. Further, as shown in the examples of FIGS. 17B and 18B in which the maximum viewing angle Φ is 179 degrees, when the maximum viewing angle Φ approaches 180 degrees, the screen diverges, and the fluoroscopy In the type virtual endoscopic image, it is not ready to discriminate what is shown. On the other hand, as shown in FIGS. 17C and 18C, in the global virtual endoscopic image, even when the maximum viewing angle Φ exceeds 180 degrees, the image is displayed without distortion. For example, a spherical polyp can be visually recognized in FIG. Therefore, for example, when the maximum viewing angle Φ of the virtual endoscopic image is continuously changed, the display means 68 is a fluoroscopic virtual endoscope when the maximum viewing angle Φ exceeds a preset threshold value. The display is changed from an image to a global virtual endoscopic image. This threshold value is a value representing the degree of distortion in a fluoroscopic virtual endoscope image and the display on the screen cannot be identified, and can be set experimentally in advance.

  Further, as shown in FIGS. 17 and 18, when the display unit 64 displays a virtual endoscopic image on the output device 26, it is generated from the three-dimensional image data f stored in the storage device 28 in advance. It is displayed in association with the tomographic image of the subject. This association can be displayed, for example, so that the cross section in the tomographic image includes the line-of-sight direction L of the virtual endoscopic image. In this way, a common part of the subject can be visually recognized simultaneously by different images displayed by a plurality of methods.

  Returning to FIG. 2, the contrast display means 70 is generated by the global virtual endoscope image generated by the global image generation means 58 of the virtual endoscope image generation means 56 and the perspective image generation means 62. The perspective virtual endoscopic image is displayed in a comparable manner. At this time, the global virtual endoscopic image and the perspective virtual endoscopic image can both be virtual endoscopic images from a common viewpoint position p. In this way, it is possible to simultaneously refer to virtual endoscopic images generated by different methods from a common viewpoint position. That is, a perspective virtual endoscopic image close to an actual endoscopic image and the maximum viewing angle at the same viewpoint position p as the perspective virtual endoscopic image or a specific angle range SR is enlarged and displayed. The global virtual endoscopic image can be referred to at the same time so as to be comparable.

  FIG. 19 is a diagram illustrating an example of a virtual endoscopic image displayed on the output device 26, and a perspective virtual endoscopic image and a global virtual endoscopic image are displayed by the contrast display unit 70 of the display unit 64. It is a figure explaining the case where and are displayed so that contrast is possible. In the image of FIG. 19, a first virtual endoscopic image display area 114 and a second virtual endoscopic image display area 116 are provided, and the first virtual endoscopic image display area 114 has a fluoroscopic virtual endoscope. A mirror image is displayed, and a global virtual endoscopic image is displayed in the second virtual endoscopic image display area 116. In the example of FIG. 19, the fluoroscopic virtual endoscope image and the global virtual endoscopic image are images at a common viewpoint position p, and the maximum viewing angle Φ is different from each other. However, it is not limited to such conditions.

  In the example of FIG. 19, a third display area 118 is provided in addition to the first virtual endoscope image display area 114 and the second virtual endoscope image display area 116, and the first virtual endoscope image display area 116 is provided. A plurality of types of tomographic images corresponding to the viewpoint position p of the virtual endoscopic image displayed in the image display area 114 and the second virtual endoscopic image display area 116 are displayed. In this way, the viewpoint position p and line-of-sight direction L of the virtual endoscopic image and the tomographic image generated based on the three-dimensional image data f can be displayed in association with each other. However, the third display area 118 is not essential.

  FIG. 6 is a flowchart for explaining a main part of the control operation in the global image generation means 58 of the virtual endoscope image generation means 56. The flowchart of FIG. 6 is repeatedly executed at predetermined intervals. First, in a step (hereinafter, “step” is omitted) S1 corresponding to the storage means 28, the three-dimensional image data f of the subject is read.

  Subsequently, in S2 corresponding to the viewpoint position setting means 50, the visual field setting means 52, and the like, the values of the viewpoint position p, the line-of-sight direction L, and the maximum visual field angle Φ are set, and the view that becomes the reference of the direction in the subsequent processing. An axial direction vector d_v, a rightward direction vector d_r, and an upward direction vector d_u are set. At the time of generating a virtual endoscopic image, an opacity table is set for representing the opacity of a voxel in the volume rendering method. This opacity table associates, for example, the three-dimensional image data f, which is a CT value taken in S1, with the opacity in volume rendering. This opacity table may be set and stored in advance, or may be set each time so that the user can change contrast and transparency.

  The visual axis direction vector d_v, the rightward direction vector d_r, and the upward direction vector d_u will be described with reference to FIG. As shown in FIG. 7, the visual axis direction vector d_v is a direction vector that intersects the center of the projection plane 104 perpendicularly to the projection plane 104, and corresponds to the line-of-sight direction L set by the viewpoint position setting means 50. . Further, the right direction vector d_r and the upward direction vector d_u define the right direction and the upward direction on the projection plane 104, and are vectors orthogonal to each other in the projection plane 104. Specifically, in this embodiment, the right direction vector d_r is defined as a vector parallel to the long side direction of the projection plane 104 and the upward direction vector d_u is defined as a vector parallel to the short side direction of the projection plane 104. Note that the projection plane 104 is a contact surface of the virtual projection plane 102 in the periphery intersecting the visual axis direction vector d_v in FIG.

  Returning to FIG. 6, in S <b> 3, coordinates (x, y) representing the position on the projection plane 104 on the image are initialized to (x, y) = (0.0). In the present embodiment, the coordinate (x, y) on the projection plane 104 is the center of the projection plane 104, that is, the intersection of the visual axis direction vector d_v (gaze direction L) and the coordinate (x, y) plane 104 is the coordinate origin. It is said. Further, a standardized coordinate system is used in which the length of the long side of the projection surface 104 in the positive direction and the length in the negative direction, and the length of the short side in the positive direction and the length in the negative direction are normalized to 1, respectively. It is done. Therefore, the coordinates (x, y) are −1 ≦ x, y ≦ 1. An example of this normalized coordinate system is shown in FIG. This coordinate (x, y) represents each pixel on the projection plane 104.

  In S4, a ray direction calculation subroutine for calculating the ray direction corresponding to the pixel on one of the pixels on the projection plane 104 is executed. FIG. 8 is a flowchart for explaining the outline of the control operation in this ray direction calculation subroutine. First, in SA1, the coordinates (x, y) on the projection plane 104, the visual axis direction vector d_v, the rightward direction vector d_r, and the upward direction vector d_u set in S2 are read.

In SA2, the θ1 calculation subroutine for calculating the rotation angle θ1 around the upward direction vector d_u of the ray corresponding to the pixel at the coordinates (x, y) is executed. FIG. 9 is a flowchart for explaining the outline of the control operation in the θ1 calculation subroutine. First, in SB1, coordinates (x, y) on the projection plane 104 are set. Subsequently, in SB2, the distance l between the coordinates (x, y) set in SB1 and the origin is calculated. This distance l is a geometric distance and is calculated based on l ² = x ² + y ² .

  Subsequently, in SB3, the relationship f (l) for associating the distance l calculated in SB2 with the rotation angle θ1 around the upward vector d_u at the coordinates (x, y) is set. In this embodiment, f (l) is set so that the distance l and the rotation angle θ1 have a linear relationship as shown in FIG. When f (l) is set in this way, the coordinates (x, y) corresponding to each rotation angle θ1 on the projection plane 104 expressed in the standardized coordinate system are shown in FIG. It is. The dotted lines in FIG. 12 are lines connecting coordinates (x, y) at which the rotation angles θ1 are π / 4, π / 2, 3π / 4, and π, respectively.

  In SB4, the rotation angle θ1 around the upward direction vector d_u is calculated based on the distance l calculated in SB2 and the relationship f (l) between the distance l set in SB3 and the rotation angle θ1.

  Returning to FIG. 8, in SA3, a θ2 calculation subroutine for calculating the rotation angle θ2 around the rightward direction vector d_r of the ray corresponding to the pixel at the coordinates (x, y) is executed. FIG. 10 is a flowchart for explaining the outline of the control operation in the θ2 calculation subroutine. First, in SC1, coordinates (x, y) on the projection plane 104 are set. Subsequently, in SC2, it is determined whether or not the value of the x coordinate is zero. If the value of the x coordinate is 0, the determination in this step is affirmed and SC3 is executed. If the value of the x coordinate is not 0, the determination at this step is denied and SC7 is executed.

  In SC3 executed when the value of the x coordinate is 0, the rotation angle θ2 is once set to 0. Subsequently, in SC4, it is determined whether or not the y coordinate is a value greater than zero. If the y coordinate is greater than 0, the determination in this step is affirmed and SC5 is executed. If the y-coordinate value is 0 or less, the determination at step SC4 is negative and SC6 is executed. In SC5, the value of the rotation angle θ2 that was set to 0 in SC3 is set to π / 2. In SC6, the value of the rotation angle θ2 that was set to 0 in SC3 is set to −π / 2.

  In SC7 executed when the value of the x coordinate is not 0, the value of the rotation angle θ2 is set as an arctangent of the value of the x coordinate with respect to the value of the y coordinate, that is, θ2 = atan (y / x). .

  In SC8, it is determined whether or not the value of the x coordinate is less than 0 and the value of the y coordinate is greater than 0. If this determination is affirmative, SC9 is executed. In SC9, π is added to the value of the rotation angle θ2 calculated in SC7.

  In SC10, it is determined whether or not the value of the x coordinate is 0 or less and the value of the y coordinate is 0 or less. If this determination is affirmative, SC10 is executed. In SC10, π is added to the value of the rotation angle θ2 calculated in SC7.

  In SC12, the value of θ2 obtained in the execution of SC1 to SC11 is set as the rotation angle θ2 around the rightward direction vector d_r. Note that θ2 set in this way corresponds to an angle formed by a straight line connecting the coordinates (x, y) and the origin and the x axis, as shown in FIG.

  Returning to FIG. 8, in SA4, the visual axis direction vector d_v is rotated around the upward direction vector d_u by θ1 to be d′ _ray. In SA5, the vector d′ _ray calculated in SA4 is rotated by θ2 around the rightward direction vector d_r to be d ″ _ray. In SA6, the vector d ″ _ray calculated in SA5 is determined as a vector d_ray indicating the direction of the ray corresponding to the pixel at the coordinates (x, y).

  Returning to FIG. 6, in S5, a shadow value calculation subroutine for calculating the shadow value of the pixel corresponding to the coordinates (x, y) is executed. FIG. 11 is a flowchart for explaining the outline of the control operation in this shadow value calculation subroutine.

  First, in SD1, three-dimensional image data f, a viewpoint position p, a ray direction vector d_ray, and an opacity table are set. Among these, the three-dimensional image data f is taken in in S1 of FIG. 6, and the viewpoint position p and the opacity table are set in S2. The ray direction vector d_ray is calculated in S4.

  In SD2, the intersection start position index k_begin at which the ray starts to intersect with the three-dimensional image and the intersection end position index k_end at which the ray ends the intersection with the three-dimensional image are determined for the position where the ray and the three-dimensional image intersect. Each is calculated. The intersection of the ray and the three-dimensional image indicates that the ray enters the volume region of the subject in the three-dimensional image when the ray advances toward the ray direction vector d_ray that is the traveling direction. Specifically, when a point on a ray is represented by p + k * d_ray using a viewpoint position p, a unit length ray direction vector d_ray, and an arbitrary real position index k, the ray is 3 The value of the position index k at the start of the intersection with the dimensional image is the intersection start position index k_begin, and the value of the position index k at the end of the intersection is the intersection end position index k_end.

  FIG. 14 is a diagram for explaining the intersection start position and the intersection end position. In FIG. 14, attention is paid to a ray corresponding to a certain pixel on the projection plane from the viewpoint position p. This ray enters the inside of the three-dimensional image corresponding to the subject represented in a cubic shape at point A, and goes out of the three-dimensional image at point B. This point A corresponds to the intersection start position, and point B corresponds to the intersection end position.

  In SD3, the value of the position index k for representing the position on the ray is set as the intersection start position k_begin.

  In SD4, a target three-dimensional pixel position v representing a target position in a three-dimensional image including the subject is calculated. The target three-dimensional pixel position v is represented by v = p + k * d_ray using the position index k set in SD3.

  In SD5, the opacity at the target three-dimensional pixel position v calculated in SD4 is calculated. In this embodiment, the opacity is specifically converted from the pixel value stored in the storage means 28 as the three-dimensional image data f using the opacity table. Specifically, for example, pixel values of points having predetermined intervals in three axial directions orthogonal to each other in advance in the subject are stored in advance as three-dimensional image data f. In this case, pixel values of eight neighboring points surrounding the target three-dimensional pixel position v are extracted, and the opacity obtained from the pixel values of the eight neighboring points is set as the opacity of each of the eight points and the target three-dimensional pixel value. Linear interpolation according to the distance from the pixel position v is performed, and the opacity at the target three-dimensional pixel position v is calculated. In the example of FIG. 14 described above, each point represented by a triangle between the point A as the intersection start position and the point B as the intersection end position corresponds to the target three-dimensional pixel position v.

The calculation of the opacity will be described in detail with reference to FIG. The XYZ coordinate system in FIG. 15 is shown for convenience of explanation, and may be different from the coordinate system in the above-described embodiment. In FIG. 15, sample points S (p, q, r) correspond to the target three-dimensional pixel position v for which opacity is to be calculated. The lattice point Pi represents a point where opacity is obtained in advance in f in the three-dimensional image. The opacity α _S at the sample point S (p, q, r) in FIG. 15 is the following linearity from the opacity α _Pi (i = 1, 2,. Calculated by complementation.
α _s = ω ₁ * α _P1 + ω ₂ * α _P2 + ... ω ₈ * α _P8
Here, ω _i represents a coefficient and is defined by the following equation.
ω ₁ = {1− (p−x)} {1− (q−y)} {1− (r−z)}
ω ₂ = (p−x) {1− (q−y)} {1− (r−z)}
ω ₃ = {1− (p−x)} (q−y) {1− (r−z)}
ω ₄ = (p−x) (q−y) {1− (r−z)}
ω ₅ = {1− (p−x)} {1− (q−y)} (r−z)
ω ₆ = (p−x) {1− (q−y)} (r−z)
ω ₇ = {1- (p−x)} (q−y) (r−z)
ω ₈ = (p−x) (q−y) (r−z)

In SD6, a gradient vector (gradient vector) grad (v) is calculated at eight neighboring points surrounding the target three-dimensional pixel position v. This gradient vector grad (v) is a pixel value f _u for pixels adjacent in the direction of the line-of-sight vector d_v, the upward direction vector d_u, and the rightward direction vector d_r for each of the eight neighboring points surrounding the target three-dimensional pixel position v _{. It} represents changes in _{v and r} , and is expressed as a vector whose component is the amount of change in each direction of the line-of-sight direction vector d_v, the upward direction vector d_u, and the rightward direction vector d_r. Specifically, for example,
grad (v) = ( _{fv, u, r-} fv _{-1, u, r} , _{fv, u, r-} fv _{, u-1, r} , _{fv, u, r-} fv _{, u, r-1} )
It is expressed as Here, f _{v, u, r} represents a pixel value at the target three-dimensional pixel position v, and f _{v-1, u, r} is a pixel position positioned in the d_v direction before the target three-dimensional pixel position v. Represents the pixel value. The same applies to the upward direction and the right direction.

  In SD7, a shadow value at the target three-dimensional pixel position v is calculated. Also in the calculation of the shadow value, similarly to the calculation of the opacity in SD5 described above, the shadow values of the eight neighboring points surrounding the target three-dimensional pixel position v are calculated, and the respective shadow values of the eight neighboring points are calculated as 8 By performing linear interpolation according to the distance between each of the points and the target 3D pixel position v, a shadow value at the target 3D pixel position v is calculated.

The calculation of the shadow value will also be described with reference to FIG. In FIG. 15, a sample point S (p, q, r) corresponds to a target three-dimensional pixel position v for which a shadow value is to be calculated. The lattice point Pi represents a point where opacity is obtained in advance in f in the three-dimensional image. The shadow value c _s ^k (k = R, G, B, respectively) at the sample point S (p, q, r) in FIG. 15 is the shadow value c _Pi ^k (i = 1, 2, ...) are calculated by the following linear storage.
c _s ^k = ω ₁ * c _P1 ^k + ω ₂ * c _P2 ^k +... + ω ₈ * c _P8 ^k
Note that ωi is the same as the aforementioned coefficient. Here, _{c Pi} ^k in Pi is obtained by multiplying the color information _{I Pi} ^k at Pi in the inner product value of the visual axis direction vector d_v the normal vector _{N Pi} as follows. That is,
c _Pi ^k = (I · N _Pi ) · I _Pi ^k
It is. Here, N _Pi is obtained by normalizing the gradient vector grad (Pi) at the eight neighboring points Pi surrounding the target three-dimensional pixel position v calculated in SD6, and is obtained by the following equation.
NPi = grad (Pi) / || grad (Pi) ||

Subsequently, in SD8, shading values are accumulated. That is, from k = k_begin corresponding to the intersection start position index, the shadow value of the three-dimensional image overlapping the ray is integrated. When Ck is an integrated value of the shadow value up to the sample point k, Ck is expressed by the following equation.
Ck = Ck-1 + βk * αk * ck
Here, ck is the shadow value at the sample point k, and αk is the opacity at the sample point k. Βk is an integrated value of transparency up to the sample point k and is defined by the following equation.
βk = βk-1 * (1-αk)
Note that β0 which is an initial value of β is set to β0 = 1.

  In SD9, since the target three-dimensional pixel position v is advanced by a predetermined direction in the ray direction vector d_ray direction on the ray, the value of the position index k is increased by a predetermined value step.

  In SD10, it is determined whether a predetermined stop condition for ending this subroutine is satisfied, or whether the value of the position index is greater than or equal to the intersection end point k_end. Among these, the predetermined cutoff condition is, for example, that the integrated value of the shadow value calculated in SD8 exceeds a predetermined value corresponding to opaqueness. In this way, when the ray casting is sequentially performed on the pixels on the projection plane and the sample point close to the projection plane is opaque, the shadow value of the sample point behind the sample point does not affect the projection plane. Therefore, there is an advantage that the calculation can be omitted. In addition, when the position index value is equal to or greater than the intersection end point k_end, the ray has not already overlapped with the three-dimensional image, so that it is not necessary to accumulate shadow values. As described above, when the predetermined stop condition is satisfied or the value of the position index crosses the end point k_end or more, the determination of this step is affirmed and the shadow value calculated in the immediately preceding SD8 is determined. The integrated value is output and this flowchart is terminated. The integrated value of the shadow values calculated in this way is used as a shadow value corresponding to each pixel on the projection plane 104. On the other hand, when the determination of SD10 is negative, the processes of SD4 to SD10 are repeatedly executed for the target three-dimensional pixel position v newly set in SD9.

  Returning to FIG. 6, in S <b> 6, the coordinates (x, y) corresponding to the pixel on the projection plane 104 are updated in order to calculate the shadow value for another pixel on the projection plane 104. Specifically, for example, the coordinates (x, y) are set so as to focus on the pixel located one right, and in the case where there is no further pixel on the right, focus on the pixel located on the leftmost one below. Is updated.

  In S7, it is determined whether or not the calculation of the shadow value has been completed for all the pixels on the projection plane 104. When the calculation of the shadow value has been completed for all the pixels on the projection plane 104, the determination in this step is affirmed and S8 is executed. If there is a pixel whose shadow value has not been calculated on the projection plane 104, the determination in this step is denied, and the shadow value is calculated for the pixel corresponding to the coordinate (x, y) updated in S6. Is done.

  In S8 executed when the determination in S7 is affirmed, a global virtual endoscopic image is generated. Specifically, the color tone of each pixel on the projection surface 104 is a shadow value obtained by repeatedly executing S4 to S7, thereby generating a global virtual endoscope image. S9 executed subsequent to S8 corresponds to the display means 64, and the generated virtual endoscopic image is output to the output device 26, that is, displayed.

  FIG. 20 shows a view angle display control in which when a virtual endoscopic image is displayed on the output device 26 in S8, a symbol for displaying a view angle in the image is superimposed on the virtual endoscopic image. It is a flowchart explaining a control action. The control operation in this flowchart corresponds to the display control means 68.

  First, in SE1, it is determined whether or not it is set by the user to superimpose symbols for displaying the viewing angle. If the user inputs to display, the determination in this step is affirmed, and SE2 and subsequent steps are executed. If the user does not input to display, the determination in this step is denied and the flowchart is terminated.

  In SE2, on the projection planes 104 and 106 of the virtual endoscopic image, a line connecting pixels corresponding to the viewing angle to be displayed is calculated. In the example of FIG. 16 described above, for example, a circle connecting pixels corresponding to a viewing angle of 180 degrees and a circle connecting pixels corresponding to a viewing angle of 320 degrees are calculated.

  In SE3, the display is performed with the line calculated in SE2 superimposed on the virtual endoscopic image. Specifically, the line calculated in SE2 is displayed in a color that can be identified in the virtual endoscopic image. In the example of FIG. 16, it is displayed by a white line.

  FIG. 21 shows a global virtual endoscope image generated by the global image generation means 58 of the virtual endoscope image generation means 56 in S8 and a perspective virtual endoscope generated by the perspective image generation means 62. It is a flowchart explaining the control action | operation in the display switching control for switching and displaying an image. The control operation in this flowchart corresponds to the display switching means 68.

  First, in SF1, it is determined whether or not the displayed virtual endoscopic image is a fluoroscopic virtual endoscopic image. When the displayed virtual endoscopic image is a fluoroscopic virtual endoscopic image, the determination in this step is affirmed and SF2 is executed. On the other hand, if the displayed virtual endoscopic image is a global virtual endoscopic image, the determination in this step is denied and SF5 is executed.

  In SF2, it is determined whether or not an operation for switching the display has been performed by the user. When the user performs an operation for switching the display, the determination in this step is affirmed and SF4 is executed. If the user has not performed an operation for switching the display, the determination at this step is denied and SF3 is executed.

  In SF3, it is determined whether or not the viewing angle Φ of the displayed perspective virtual endoscope image exceeds a predetermined threshold value. The threshold value of the viewing angle is an angle that is distorted to such an extent that an object displayed in the image cannot be identified when a perspective virtual endoscopic image is used. If the viewing angle Φ exceeds a predetermined threshold, the determination in this step is affirmed and SF4 is executed. On the other hand, when the viewing angle Φ does not exceed a predetermined threshold value, this flowchart is terminated without switching the display.

  In SF4 executed when the determination of SF2 is affirmed and when the determination of SF3 is affirmed, the virtual endoscopic image displayed by the display unit 64 is changed from the fluoroscopic virtual endoscopic image to the global virtual image. Switch to endoscopic image.

  In SF5, it is determined whether or not an operation for switching the display has been performed by the user. If the user performs an operation for switching the display, the determination in this step is affirmed and SF6 is executed. If the user has not performed an operation for switching the display, the determination in this step is denied, and the flowchart is terminated without switching the display.

  In SF6 executed when the determination in SF5 is affirmed, the virtual endoscopic image displayed by the display unit 64 is switched from the global virtual endoscopic image to the perspective virtual endoscopic image.

  According to the above-described embodiment, a global virtual endoscopic image having a field of view surrounding the viewpoint position p in a spherical shape is generated by the global image generating unit 58 of the virtual endoscopic image generating unit 56, and is displayed by the display unit 64. Since the global virtual endoscopic image is displayed on the output device 26, the virtual endoscopic device 10 can obtain a global virtual endoscopic image having a wide field of view surrounding the viewpoint position p in a spherical shape.

  According to the above-described embodiment, the global image generation unit 58 of the virtual endoscopic image generation unit 56 generates the ray from the viewpoint position p to the field of view and uses the volume rendering method to generate the global virtual endoscope. Since the mirror image is generated, a global virtual endoscopic image having a wide field of view surrounding the viewpoint position p in a spherical shape can be generated based on the three-dimensional image data f.

  According to the above-described embodiment, the global image generating unit 58 of the virtual endoscopic image generating unit 56 generates a global virtual endoscopic image having a preset maximum viewing angle Φ. A spherical virtual endoscopic image having a wide viewing angle that is freely set in advance by the operator can be obtained with respect to the spherically surrounding visual field. In particular, it is possible to generate a global virtual endoscopic image having a maximum viewing angle Φ exceeding 180 degrees, which cannot be realized by an actual endoscope apparatus or fluoroscopy. In addition, even if the maximum viewing angle Φ is less than 180 degrees, a global virtual endoscopic image corresponding to the maximum viewing angle Φ that diverges and is difficult to interpret in a fluoroscopic virtual endoscopic image is generated. be able to.

  According to the above-described embodiment, the display unit 64 superimposes the global virtual endoscopic image and the symbol for displaying the viewing angle on the output device 26 so that the viewpoint position p is spherically surrounded. A viewing angle can be grasped in a global virtual endoscopic image having a visual field.

  According to the above-described embodiment, in the virtual endoscopic image generation unit 56, in addition to the global virtual endoscopic image generated by the global image generation unit 58, the viewpoint position p is generated by the perspective image generation unit 62. A fluoroscopic virtual endoscopic image generated by fluoroscopy is generated from the image, and the global virtual endoscopic image and the fluoroscopic virtual endoscopic image are switched and displayed by the display switching unit 68 of the display unit 64. Therefore, by switching between the global virtual endoscopic image having a wide field of view that surrounds the viewpoint position spherically at the same viewpoint position p, and the perspective virtual endoscopic image close to the actual endoscopic image. Can be observed.

  Further, according to the above-described embodiment, the display means 64 switches between the global virtual endoscopic image and the fluoroscopic virtual endoscopic image in accordance with the maximum viewing angle Φ and displays them on the output device 26, so the maximum viewing angle. The virtual endoscopic image can be automatically switched and displayed according to Φ.

  According to the above-described embodiment, in the virtual endoscopic image generation unit 56, in addition to the global virtual endoscopic image generated by the global image generation unit 58, the viewpoint position p is generated by the perspective image generation unit 62. A perspective virtual endoscopic image generated by a fluoroscopy method is generated from the image, and the global virtual endoscopic image can be compared with the fluoroscopic virtual endoscopic image by the contrast display means 70 of the display means 64 Therefore, the virtual endoscopic image generated by a different method from the common viewpoint position can be referred to at the same time, and the convenience in observing the virtual endoscopic image is improved.

  Subsequently, another embodiment of the present invention will be described. In the following description, portions common to the embodiments are denoted by the same reference numerals and description thereof is omitted.

  This embodiment relates to another aspect in the case where the global image generation means 58 of the virtual endoscope image generation means 56 generates a global virtual endoscope image in the above-described embodiment. When the global image generation unit 58 generates the global virtual endoscopic image, the ray direction is calculated (S4 in FIG. 6). At this time, the rotation angle θ1 around the upward direction vector d_u is calculated. (SA2 in FIG. 8, FIG. 9). Here, the rotation angle θ1 is calculated based on the coordinate (x, y) on the projection plane 104, the distance l between the coordinate origins, and a predetermined function θ1 = f (l). In the above-described embodiment, f (l) is a linear function with respect to l (see FIG. 13). According to this aspect, as shown in FIG. 12, for each pixel on the standardized projection plane 104, the distance from the origin and the viewing angle corresponding to the pixel have a linear relationship.

  On the other hand, in this embodiment, f (l) is a non-linear function with respect to l as shown in FIG. This f (l) is a function in which f (l) with respect to the increment of l, that is, the increment of θ1, decreases as l increases. Therefore, in the state close to the edge of the image, that is, when l is large, the increment of θ1 with respect to the increment of l is decreased. FIG. 22 is a diagram corresponding to FIG. 12 of the above-described embodiment, and among the pixels on the projection surface 104, a line connecting pixels corresponding to the same viewing angle is displayed by a dotted line. In other words, the center portion of the image is expanded and displayed, while the end portion of the image, that is, the viewing angle, is compressed and displayed.

  As described above, according to the global image generation unit 58 of the present embodiment, for each pixel on the projection surface 104, the viewing angle corresponding to the pixel is set to each pixel on the standardized projection surface 104 and the origin. It is possible to have a non-linear relationship with respect to the distance. At this time, the function f (l) is not limited to that shown in FIG. That is, in this way, the function f (l) is not limited to a linear one, and the global virtual endoscopic image generated by the selection can be changed. It is also possible to switch f (l) based on the user's operation of the input device 24 or the like.

  In the present embodiment, the virtual endoscope apparatus 10 includes a specific area setting unit 54, and the global image generation unit 58 of the virtual endoscopic image generation unit 56 includes the specific area processing unit 60. Have.

  The specific area setting unit 54 sets an area having a predetermined viewing angle as the specific area based on, for example, an operation of the input device 24 by the user, and should display the specific area, that is, the specific area should be enlarged. Set whether to display in a reduced size.

  Further, the specific area processing means 60 performs processing for displaying the area of the viewing angle that has been set as the specific area by the specific area setting means 54 by the set display method of the specific area. Specifically, the function f (l) that can display the region of the viewing angle that has been set as the specific region by the specific region setting means 54 by the display method of the set specific region is selected. As shown in the above-described second embodiment, the steeper gradient of f (l) with respect to l is enlarged in the global virtual endoscopic image, and is reduced and displayed as the gradient is gentler. Therefore, when it is set by the specific area setting unit 54 to display an enlarged (expanded) area of a specific viewing angle, for example, “range of 60 to 80 degrees”, the area of the viewing angle is displayed. In the corresponding l range, f (l) is set such that the gradient of f (l) with respect to l becomes steeper than the l range corresponding to the region other than the specific viewing angle. Further, when it is set by the specific region setting means 54 that the region of the specific viewing angle is to be reduced (compressed), the f (l) with respect to l in the range of l corresponding to the region of the viewing angle. F (l) is set such that the gradient is gentler than the range of l corresponding to the region other than the specific viewing angle. For this setting, a function most suitable for realizing the setting in the specific area setting means 54 may be selected from a plurality of types of functions f (l) prepared in advance, or the specific area setting means A function f (l) suitable for realizing the setting in 54 may be generated.

  Further, instead of designating the region of the viewing angle to be enlarged or reduced as the specific region setting means 54, the function f (l) can be generated by the user.

  FIG. 24 is a diagram illustrating an example of an interface provided when f (l) is generated by the user. This interface is displayed on the output device 26, for example, and the display is changed corresponding to the input by the input device 24. In the interface of FIG. 24, a graph corresponding to FIG. 13 or FIG. 23 of the above-described embodiment is displayed, and a plurality of handles 90 represented by squares are provided on the graph. The shape of the function f (l) can be changed by moving the handle with a mouse as the input device 24, for example. In this way, the rotation angle θ1 (= f (l)) with respect to the distance l between the coordinate (x, y) on the standardized projection plane and the origin can be set intuitively or by trial and error.

  According to the above-described third embodiment, the global image generation unit 58 of the virtual endoscope image generation unit 56 generates a global virtual endoscope image in which a specific angle region is expanded or compressed at the viewing angle. Therefore, it is possible to obtain a global virtual endoscopic image that has a wide field of view that surrounds the viewpoint position p in a spherical shape and that is enlarged or reduced with respect to a specific viewing angle range.

  According to the third embodiment described above, the position and range of the specific angle area can be changed by the specific area setting unit 54. For example, the angle can be specified by designating a specific angle area that the operator wants to enlarge and display. A global virtual endoscopic image with an enlarged region can be obtained.

  In the present embodiment, the virtual endoscope apparatus 10 includes a feature part extraction unit 72. The feature part extraction unit 72 extracts a part to be noted as a feature part in the subject to be imaged by the virtual endoscope apparatus 10. The site to be noted corresponds to, for example, a tumor when the virtual endoscope device is used for a virtual endoscopic image of the large intestine. Specifically, the feature part extraction unit 72 detects the position of the part to be noted based on the three-dimensional image data f stored in the storage device 28 in advance. For this detection, for example, a feature appearing in the three-dimensional image data corresponding to the region to be noted is stored in advance, and the three-dimensional image data f includes the feature, that is, matches the feature in the three-dimensional image data f. Alternatively, it may be performed by so-called model matching in which, when a similar part exists, the part is detected as a characteristic part.

  On the other hand, in the present embodiment, the specific region setting unit 54 includes the feature when the feature part extracted by the feature part extraction unit 72 is included in the global virtual endoscopic image generated by the global image generation unit 58. The specific area is set so that the range of the viewing angle including the part is enlarged and displayed.

  According to the above-described embodiment, the feature part including the preset feature is extracted from the three-dimensional image data f of the subject by the feature part extraction unit 72 and extracted to the global virtual endoscopic image. When a characteristic part is included, the specific angle area is set so as to include the characteristic part by the specific area setting unit, and set by the specific area setting unit 54 by the virtual endoscope image generation unit Since the specific angle region is enlarged and displayed by the specific region processing means 60, the whole globe has a wide field of view that surrounds the viewpoint position in a spherical shape, and a feature portion including a preset feature is enlarged and displayed. Type virtual endoscopic images can be obtained.

  At this time, when the feature part extracted by the feature part extraction means 72 is displayed on the virtual endoscopic image in the display switching means 68 of the display means 64, the global virtual endoscopic image is converted from the perspective virtual endoscopic image. You can switch to a mirror image for display. In this way, when the feature portion is displayed, the feature portion is switched to the global virtual endoscopic image that can be enlarged and displayed.

  In the above-described embodiment, the viewing angle display means 66 of the display means 64 superimposes a symbol for displaying the viewing angle in the virtual endoscopic image on the virtual endoscopic image, so that the viewing angle in the virtual endoscopic image is displayed. Was made visible. In the present embodiment, the viewing angle display means 66 makes the hue different between a region that exceeds the boundary angle that is a preset viewing angle and a region that falls below the boundary angle in the virtual endoscopic image. Thus, the viewing angle is displayed in the virtual endoscopic image.

  FIG. 25 is an image showing an example of a virtual endoscopic image in which the hue of the region exceeding the size of the boundary angle and the hue of the region below the boundary angle are made different by the viewing angle display unit 66. . FIG. 25 shows an example of a screen displayed on the output device 26, for example. In the example of FIG. 25, the boundary angle is, for example, a viewing angle of 180 degrees. In the region 110 where the viewing angle exceeds 180 degrees which is the boundary angle, the hue is displayed darker than in the region 111 where the viewing angle is less than 180 degrees which is the boundary angle. At this time, the viewing angle at which the boundary 112 between the region 110 and the region 111 is the boundary angle corresponds to 180 degrees.

  According to the above-described embodiment, the hue of the region 110 whose viewing angle is 180 degrees or more, which is the boundary angle, is different from the hue 111 of the region whose viewing angle is less than 180 degrees, which is the boundary angle, and is displayed so as to be dark. Therefore, in the virtual endoscopic image, it is possible to distinguish between a region having a viewing angle narrower than a preset boundary angle and a region having a viewing angle wider than the boundary angle. In addition, the boundary angle can be visually recognized by the boundary 112 between the region 110 and the region 111 in the virtual endoscopic image.

  Further, according to the above-described embodiment, since the boundary angle is set to a viewing angle of 180 degrees, the region 110 having a viewing angle of 180 degrees or more, which is the boundary angle, is determined from the viewpoint position in the global virtual endoscopic image. Corresponds to the rear, and the region 111 having a viewing angle of less than 180 degrees corresponds to the front of the viewpoint position in the global virtual endoscopic image. Then, the hue of the area 110 and the hue of the area 111 are displayed differently. Therefore, in the global virtual endoscopic image, an area corresponding to the rear of the viewpoint position that could not be seen in an actual endoscopic image or in a fluoroscopic endoscopic endoscope that has been widely used in the past has been used. Can be distinguished

  As mentioned above, although the Example of this invention was described in detail based on drawing, this invention is applied also in another aspect.

  For example, in addition to the above-described embodiment, the viewpoint position p can be continuously changed based on a user operation or the like by the input device 24. Specifically, when the virtual endoscopic image generation unit 56 continuously changes the viewpoint position p by repeatedly executing the flowchart of FIG. 6, the viewpoint position p set in S2 is changed. Since a global virtual endoscopic image is continuously generated, it is possible to obtain a global virtual endoscopic image having a wide field of view that surrounds the viewpoint position in a spherical shape corresponding to the viewpoint position that is continuously changed. . That is, a virtual endoscopic image corresponding to the entry or exit of the virtual endoscope into the subject is obtained.

  In addition to the above-described embodiment, the maximum viewing angle Φ of the global virtual endoscope can be continuously changed based on the user's operation using the input device 24. Specifically, in the virtual endoscopic image generation means 56, as the maximum viewing angle Φ is continuously changed, the shape of the virtual projection plane 102 is changed so that a global virtual endoscopic image is obtained. Since it is continuously generated, it is possible to obtain a global virtual endoscopic image having a wide field of view surrounding the viewpoint position in a spherical shape corresponding to the viewpoint position that is continuously changed. In this way, virtual endoscopic images in which the maximum viewing angle continuously changes at the same viewpoint position are sequentially obtained.

  In the fifth embodiment described above, the viewing angle display means 66 of the display means 64 has a viewing angle 180 that is a boundary angle when the maximum viewing angle Φ of the generated global virtual endoscopic image is 180 degrees or more. For the area above the degree, a global virtual endoscopic image having a hue different from that of the area where the viewing angle is less than the boundary angle, specifically, the lightness of darkness, was generated. Absent. For example, a region where the viewing angle is within the boundary angle is displayed in color (colored), while a region where the viewing angle is larger than the boundary angle can be displayed in black and white. The boundary angle is not limited to 180 degrees, and can be set to an arbitrary angle in the global virtual endoscopic image.

  In the generation of the global virtual endoscope image by the global image generation means 58 of the virtual endoscope image generation means 56, the setting of the direction of the ray at the time of executing the ray casting method is described in the above embodiment. It is not limited to the method. For example, the ray corresponding to each pixel on the projection plane 104 is calculated in advance by an angle θ1 formed by the ray defining the ray and the line-of-sight direction L and a rotation angle θ2 around the line-of-sight direction L of the ray, Ray casting may be executed for each of the calculated rays.

  In the above-described embodiment, the maximum viewing angle Φ of the global virtual endoscopic image generated by the global image generating unit 58 of the virtual endoscopic image generating unit 56 is 360 degrees or less. The viewing angle Φ may exceed 360 degrees. Even when the maximum viewing angle Φ is set to 360 degrees, when the virtual endoscopic image is displayed in a square shape such as a rectangle on the output device 26, one virtual endoscopic image surrounding the viewpoint position in a spherical shape is displayed. The part may be missing and not displayed. In such a case, by setting the maximum viewing angle Φ to a value exceeding 360 degrees, all of the virtual endoscopic images that surround the viewpoint position in a spherical shape can be displayed on the output device 26. When the maximum viewing angle Φ is set to a value exceeding 360 degrees, a common region is displayed at a pair of end portions of the virtual endoscopic image that is symmetric about the center of the virtual endoscopic image (gaze direction L). The Rukoto.

  In the above-described embodiment, an example in which the maximum viewing angle Φ is 180 degrees or more is shown as the global virtual endoscopic image, but the present invention is not limited to this. That is, the virtual endoscopic apparatus of the present invention can generate a global virtual endoscopic image when the maximum viewing angle Φ is greater than 0 degrees, and even in such a case, the global virtual endoscope in the above-described embodiment can be used. The effect of the endoscopic image can be obtained similarly.

  In addition, although not illustrated one by one, the present invention is implemented with various modifications within a range not departing from the gist thereof.

10: Virtual endoscope device 54: Specific region setting means 56: Virtual endoscope image generating means 58: Global image generating means 62: Perspective image generating means 64: Display means 72: Feature part extracting means f: Three-dimensional Image data p: viewpoint position

Claims (10)

Virtual endoscopic image generation means for virtually generating a virtual endoscopic image obtained when an endoscope is inserted into the subject based on three-dimensional image data of the subject, and at least the virtual interior A virtual endoscopic device having display means for displaying a virtual endoscopic image generated by an endoscopic image generation means,
The virtual endoscopic image generating means generates a global virtual endoscopic image having a field of view surrounding a viewpoint position in a spherical shape.   2. The virtual endoscopic image generation unit generates the global virtual endoscopic image using a volume rendering method by generating a ray from the viewpoint position to the visual field. The virtual endoscopic device described   The virtual endoscope apparatus according to claim 1, wherein the virtual endoscope image generation unit generates a global virtual endoscope image having a preset maximum viewing angle.   4. The virtual endoscopic image generation unit generates the global virtual endoscopic image obtained by expanding or compressing a specific angular region at the viewing angle. The virtual endoscope apparatus described in 1.   The virtual endoscope apparatus according to claim 4, wherein a position and a range of the specific angle region can be changed. Feature part extraction means for extracting a feature part including a preset feature from the three-dimensional image data of the subject;
A specific region setting unit configured to set the specific angle region so as to include the feature part when the feature part extracted by the feature part detection unit is included in the global virtual endoscopic image; ,
The virtual endoscope apparatus according to claim 4 or 5, wherein the virtual endoscopic image generation means enlarges and displays the specific angle area set by the specific area setting means.   The virtual endoscopic image generation means can continuously generate the global virtual endoscopic image in response to continuously changing the viewpoint position. The virtual endoscope apparatus described in 1.   The virtual endoscope according to any one of claims 1 to 7, wherein the display unit displays the global virtual endoscopic image and a symbol for displaying a viewing angle in an overlapping manner. apparatus. The virtual endoscope image generation means can generate a perspective virtual endoscope image generated by a perspective method from the viewpoint position in addition to the global virtual endoscope image,
9. The virtual endoscope according to claim 1, wherein the display unit switches between the global virtual endoscopic image and the fluoroscopic virtual endoscopic image for display. 10. Mirror device. The virtual endoscope image generation means can generate a perspective virtual endoscope image generated by a perspective method from the viewpoint position in addition to the global virtual endoscope image,
The virtual display according to any one of claims 1 to 8, wherein the display means displays the global virtual endoscopic image and the fluoroscopic virtual endoscopic image in a comparable manner. Endoscopic device.