KR100466409B1 - System and method for displaying a virtual endoscopy and computer-readable recording medium having virtual endoscopy displaying program recorded thereon - Google Patents

Abstract

Disclosed are a virtual endoscope system and method and a computer-readable recording medium storing the same. According to the present invention, a volume rendering image corresponding to a region to be observed from three-dimensional volume data is generated, and volume rendering images, two-dimensional images, virtual endoscope images, and paths generated on a plurality of viewers divided into one screen. Each reconstruction image is displayed. The path reconstruction image displayed as the path is set by the user in the 2D image or the 3D volume rendered image is updated, and when the camera position change request is requested by the user's operation, The virtual endoscope image data adjusted accordingly is generated, and the virtual endoscope image and the 2D image are updated based on the generated virtual endoscope image data.

As a result, the image generated by volume rendering is used as a reference image without preprocessing 3D volume data, and the user can easily set and modify a path by using the reference image, and the direction of the camera is changed to 3D in 2D. You can adjust the dimensions.

Description

System and method for displaying a virtual endoscopy and computer-readable recording medium having virtual endoscopy displaying program recorded thereon}

The present invention relates to a virtual endoscope system, and more particularly, a virtual endoscope system and a virtual endoscope display method capable of setting a path using a cross-sectional image and a volume rendering image of a virtual endoscope, and a program for performing the method on a computer. A computer readable recording medium.

In general, an image reconstruction system including a virtual endoscope system has many applications, but may be used in the medical imaging field of a human body structure having a three-dimensional shape.

In particular, the 3D medical imaging technique refers to generating a 3D image from a series of 2D medical images obtained from a computed tomography (CT) or a magnetic resonance imaging (MRI). Diagnosis using only two-dimensional images has a disadvantage in that it is difficult to obtain an overall three-dimensional effect and cannot observe an arbitrary cross section. However, the 3D medical imaging technique enables accurate location determination of the affected part and predicts the surgical method more realistically.

This conventional virtual endoscope system used a two-dimensional image to adjust the movement of the camera (or virtual camera). However, the method of using the 2D image has a problem that it is difficult to determine the location of the camera in the human body observed.

In addition, there is a problem that it is difficult to perform an operation that shows a continuously changing image while moving the camera or the endoscope, that is, automatic navigation.

With this in mind, to show the contours of the human structure, the segmentation process separates the data of one structure from the other, and then displays the entire contour with surface extraction rendering techniques, and displays the exact position of the camera. More advanced methods have been developed, and under these circumstances, algorithms have been developed to automatically find routes for navigation.

However, even with this advanced approach, fully automatic segmentation is difficult, and the more complicated the structure, the more difficult the segmentation can be done automatically.

In addition, the algorithm that automatically finds the path of the endoscope does not find the path to a certain level desired by the user, and there is a problem in that information about the structure wall is always known through segmentation.

Accordingly, the present invention has been made in an effort to solve such a conventional problem, and an object of the present invention is to perform a path setting directly on a 3D volume rendered image without a preprocessing process, thereby reducing the time for diagnosis and reconstructing a multifaceted path including a path. By displaying an image, to provide a three-dimensional image effect on a two-dimensional, to provide a virtual endoscope system for improving the efficiency of the diagnosis.

Another object of the present invention is to provide a virtual endoscope display method using the virtual endoscope system.

Still another object of the present invention is to provide a computer-readable recording medium storing a program for performing the above-described virtual endoscope display method on a computer.

1 is a view for explaining a virtual endoscope system according to an embodiment of the present invention.

2 shows an example of a screen of the virtual endoscope system according to the present invention.

3 is a flowchart illustrating a virtual endoscope display method according to the present invention.

FIG. 4 is a flowchart for explaining the above-described step S400 of FIG. 3 in more detail.

FIG. 5 is a flowchart for explaining the above-described step S420 of FIG. 4 in more detail.

FIG. 6 is a view for explaining a process of finding a three-dimensional position of an actual internal organ from a point input on a reference image in the virtual endoscope system according to the present invention.

7 is a view for explaining the generation of the cross-sectional reconstruction image including a path according to the present invention.

8 is a view for explaining the generation of the cross-sectional reconstruction image parallel or perpendicular to the path according to the present invention.

9 is a view for explaining a two-dimensional to three-dimensional camera adjustment in accordance with the present invention.

10 illustrates a 3D volume rendering image used as a reference image of the virtual endoscope system according to the present invention.

FIG. 11 illustrates an example of a screen provided by a preferred virtual endoscope system according to the present invention based on the 3D volume rendering image of FIG. 10.

FIG. 12 illustrates a screen providing a function of allowing a user to arbitrarily modify a path designated by the user once in FIG. 11 in the preferred virtual endoscope system according to the present invention.

<Description of the symbols for the main parts of the drawings>

10: input / storage unit 20: endoscope image processing unit

22: reference image processing unit 24: path conversion unit

26: multi-faceted reconstruction image synthesizer 30: display unit

40: input unit 100: 3D volume rendering image

200: two-dimensional image 300: virtual endoscope image

400: path reconstruction image

Virtual endoscope system according to one feature for realizing the above object of the present invention,

An input / storing unit for receiving the characteristic according to the position of each part of the three-dimensional structure having a predetermined characteristic expressed as a function of three-dimensional position in the form of volume data and storing it for subsequent processing;

Based on the data stored in the input storage unit, the spatial distribution of the characteristic is displayed as a 3D image and is processed to be displayed as a reference image, and an endoscope image is generated along a predetermined path displayed directly on the reference image. An endoscope image processor configured to display the same;

A display unit configured to display the reference image and to display an endoscope image along the path; And

And an input unit for allowing a user to designate the path from the reference image.

The endoscope image processor may be configured to display the 3D reference image from volume data stored in the input storage unit, and refer to a predetermined path passing through an internal structure for viewing the endoscope image through the input unit. A reference image processor which receives and processes a predetermined curve of data on the image; The position of the volume data of the voxel corresponding to the wall of the internal structure from which the endoscope image is to be viewed is input from the path input to the reference image processor. The path of the path actually passing through the internal structure is determined by the position of the voxel. A path converting unit correcting a path so that the path can be corrected; And a multi-faceted reconstruction image synthesizer configured to generate a multi-faceted reconstructed image including the path according to the corrected path obtained by the path transform unit.

In addition, the virtual endoscope display method according to one aspect for realizing another object of the present invention, the virtual endoscope display method for storing the raw image data, classify the density value of the stored raw image data to display the virtual endoscope To

(a) generating a volume rendering image corresponding to an area to be observed from the 3D volume data;

(b) displaying each of the generated volume rendering image, two-dimensional image, virtual endoscope image, and path reconstruction image on a plurality of viewers partitioning one screen;

(c) updating the display of the path reconstruction image as a path is set in one of the 2D image and the 3D volume rendering image according to a user's manipulation;

(d) checking whether to change the position of the camera according to a user's operation; And

(e) When the camera position change request is made, generate virtual endoscope image data adjusted according to the position change of the camera, and update the virtual endoscope image and the 2D image based on the generated virtual endoscope image data. It comprises a step.

Here, the camera displays the volume rendering image, the two-dimensional image, the virtual endoscope image, and the path reconstruction image by using a predetermined time vector, and uses the change time vector by adjusting the time vector. It is preferable to obtain a three-dimensional image effect on the two-dimensional image by updating the two-dimensional image, the virtual endoscope image and the path reconstruction image.

In addition, a computer-readable recording medium storing a program for performing a virtual endoscope display method on a computer according to one aspect for realizing another object of the present invention, stores the raw image data, stored raw image A computer-readable recording medium storing a program for performing a method of displaying a virtual endoscope on a computer by classifying density values of data, the method comprising:

(a) generating a volume rendering image corresponding to an area to be observed from the 3D volume data;

(b) displaying each of the generated volume rendering image, two-dimensional image, virtual endoscope image, and path reconstruction image on a plurality of viewers partitioning one screen;

(c) updating the display of the path reconstruction image as a path is set in one of the 2D image and the 3D volume rendering image according to a user's manipulation;

(d) checking whether to change the position of the camera according to a user's operation; And

(e) When the camera position change request is made, generate virtual endoscope image data adjusted according to the position change of the camera, and update the virtual endoscope image and the 2D image based on the generated virtual endoscope image data. It comprises a step.

According to a computer-readable recording medium storing such a virtual endoscope system and a display method and a program for performing the method on a computer, an image generated by volume rendering without preprocessing 3D volume data is used as a reference image. By using the reference image, a user can easily set and modify a path, and adjust the direction of the camera in two dimensions in three dimensions.

Then, embodiments will be described so that those skilled in the art can easily implement the present invention.

1 is a view for explaining a virtual endoscope system according to an embodiment of the present invention.

Referring to FIG. 1, the virtual endoscope system according to the present invention includes an input / storage unit 10, an endoscope image processing unit 20, a display unit 30, and an input unit 40.

The input / storage unit 10 receives the characteristic according to the position of each part of the three-dimensional structure having a predetermined characteristic expressed as a three-dimensional position function in the form of volume data and stores it for later processing. For example, in the case of receiving and storing volume data as a set of CT or MRI cross-sectional scanning results, it may include a connection hardware for receiving data transmission from each measuring device and a hard disk or a memory element for storing the data. In case that the measured result is already stored in another storage medium or read in the storage medium such as its own hard disk, access hardware for reading and storing data from such internal / external storage device and Obviously, it may be composed of a hard disk or a memory device for storing the same.

The endoscope image processor 20 includes a reference image processor 22, a path converter 24, and a multi-faceted reconstructed image synthesizer 26, based on data stored in the input / storage unit 10. It is processed to display the spatial distribution of the 3D image and to display it as a reference image, to generate an endoscope image along a predetermined path displayed directly on the reference image, and to display it.

In more detail, the reference image processor 22 processes the 3D reference image from the volume data stored in the input / storage unit 10 and passes through the internal structure to view the endoscope image through the input unit 40. A predetermined path is input and processed in the form of predetermined curve data on the reference image.

The path converter 24 determines a position in the volume data of the voxel corresponding to the wall of the internal structure from which the endoscope image is to be viewed from the path input to the reference image processor 22, and the path is actually inside by the location of the voxel. Correct the path to be the path through the structure.

The multi-faceted reconstructed image synthesizer 26 generates a multi-faceted reconstructed image including the above-described path according to the corrected path obtained by the path converting unit 24.

The endoscope image processing unit 20 is a means for calculating and processing volume data, and it is apparent that the endoscope image processing unit 20 may be implemented through a computer capable of performing a series of calculation processes or a computer set or interchangeable ASIC.

The display unit 30 displays the reference image, displays the endoscope image according to the path, and the input unit 40 allows the user to specify the path from the reference image.

2 shows an example of a screen of the virtual endoscope system according to the present invention.

2, the screen of the virtual endoscope system according to the present invention is a 3D Volume Rendering Image (100), 2D (Axial Image) 200, Endoscopy Image (Endoscopy Image) ( 300 and four viewers of the MPR path image 400. Volume rendering refers to a task of extracting visual image information required for various applications from volume data of a 3D space.

When the camera is set by the user's manipulation in the 3D volume rendering image 100 of the displayed screen, the 2D image 200 and the virtual endoscope image 300 are generated accordingly to update the image already displayed. In addition, as the path is set by the user's manipulation, the path reconstruction image 400 is generated in association with the path to update the image already displayed.

In addition, in the displayed two-dimensional image 200, the path reconstruction image 400 may be generated by updating the image that is already displayed by setting the camera by the user's operation, or the virtual may be set by the user's operation. The endoscope image 300 may be generated to update an image that is already displayed.

Next, an operation for displaying the various images will be described in detail.

3 is a flowchart illustrating a virtual endoscope display method according to the present invention.

Referring to FIGS. 2 and 3, first, scan data, that is, raw image data, of a human structure obtained through a device such as CT or MRI is stored (step S100).

Then, in order to see the structure to be observed according to the user's operation, a classification process such as density value classification is performed in real time, and unnecessary parts are deleted to make the observation part appear clearly. 100) (step S200).

Subsequently, the 3D volume rendering image 100, the 2D image 200, the virtual endoscope image 300, and the path reconstruction image 400 are respectively displayed on the four viewers illustrated in FIG. 2 (S300). .

Subsequently, the path setting request is input according to the user's manipulation on the 2D image or the 3D volume rendering image (step S400). Such a path setting or a path change may be performed on a 2D image, but for convenience of the user, the user may directly set the path in the form of an arbitrary straight line or a curve on the 3D volume rendering image.

In addition, a 2D point designated in the 3D volume rendering image 100 is mapped to a point of the center of the internal structure to be viewed, that is, mapped to a 3D point, thereby generating and displaying an image according to a set path (step S500). Preferably, the path reconstruction image 400 shown in FIG. 2 will be updated.

Then, it is checked whether there is a camera adjustment according to the user's operation (step S600), and if there is no camera adjustment, feedback is returned to step S500 to continue displaying the image, and if there is a camera adjustment, the virtual endoscope image adjusted by the camera. 300 is generated (step S700). At this time, the adjusted camera is a kind of virtual camera moving on the endoscope image, that is, the viewpoint of the user's eyes, and can freely adjust the viewing angle of the camera. The volume rendering image is reconstructed from the raw image data and displayed in conjunction with the free angle of the camera.

Subsequently, the 2D image 200 and the virtual endoscope image 300 are updated and displayed (step S800).

Then, it is checked whether the 3D volume rendered image 100 is remanufactured (step S900), and when it is checked that the 3D volume rendered image is remanufactured, it is fed back to step S200.

Next, a path setting method, which is one feature of the present invention, will be described in detail with reference to the accompanying drawings.

FIG. 4 is a flowchart for explaining the above-described step S400 of FIG. 3 in more detail.

Referring to FIG. 4, as points are designated for path reconstruction by a user's manipulation in a 2D image or a 3D volume rendered image (step S410), curved control points are added (step S420).

Subsequently, a path is generated using an equation of a three-dimensional curve (step S430), and the corresponding points are stored in a predetermined path data structure while going along the curve by a predetermined interval, for example, one voxel interval (step S440). ). The various points stored in the path data structure will be used as data corresponding to the path to be reset, and in response, the path reconstruction images will be updated and displayed in step S500.

In step S410 described above, if the user uses the 2D image 200 shown in FIG. 2 as a reference image to obtain a virtual endoscope image or a path reconstruction image, With the x, y) data, it is possible to recognize which voxel is present in the voxel, and thus it is possible to recognize the z coordinate value and to know the 3D coordinates (x, y, z), which are positions on the entire volume data. .

However, if a user uses a 3D image as a reference image to obtain a virtual endoscope image or a path reconstruction image, it is necessary to efficiently recognize the z value.

Next, a method of recognizing a z value when using a 3D image, which is another feature of the present invention, as a reference image will be described in detail.

FIG. 5 is a flowchart for explaining step S420 of FIG. 4 in more detail. In particular, a process of mapping a 2D point designated in a 3D volume rendering image to a center point of a structure is described, and FIG. In the virtual endoscope system, it is a view for explaining a process of finding a three-dimensional position of an actual internal organ from a point input on a reference image.

5 and 6, first, as a mouse is input according to a user's operation on a two-dimensional image displayed by 3D volume rendering (step S4210), a minimum value and a maximum value of a depth to be inspected are set (step S4210). S4220).

Then, the point in the plane coordinate system is converted into a point in the spatial coordinate system by multiplying the time inverse matrix (step S4230).

Subsequently, a predetermined ray is emitted from the set minimum depth (step S4240). Of course, the rays emitted here will be virtual straight rays emitted in a virtual three-dimensional space rather than physical rays.

Subsequently, it is checked whether a voxel corresponding to the wall of the structure has been met (step S4250), and when it is checked that the voxel corresponding to the wall of the object has been met, the location of the encounter at the involume point is recorded (step S4260). That is, the voxels lying on a straight line along a given viewing direction or camera direction in a 2D image input to the screen, that is, each unit part constituting the volume data and surrounding values are checked. Check if it corresponds to the density value of the wall.

Subsequently, it is checked whether a voxel corresponding to the wall of the object has been met (step S4270), and when it is checked that the voxel corresponding to the wall of the object has been met, the location of the encounter at the outvolume point is recorded (step S4280).

Subsequently, the midpoint result of the involume point and the outvolume point is output, and based on this, the point input in the volume rendering image may be converted to the center of the structure (step S4290).

As described above, in FIG. 5, the black voxel is a voxel having a density value of the human body structure, the first point in the direction of arrow progression indicates before entering the structure, the second point indicates the inside of the structure, and the third point. Refers to just passing through the inside of the structure.

Meanwhile, when the endoscope is diagnosed with a virtual endoscope system, it is very useful for diagnosis if there is information on surrounding structures. In particular, medical staff want a reconstructed image that includes a path through the centerline of the structure.

In response to the request of the medical staff, according to the virtual endoscope system of the present invention, a reconstructed image including a path is shown as a reference image. In addition, the image rotates 360 degrees to create an image that provides accurate information about the surrounding structures.

7 is a view for explaining the generation of a cross-sectional reconstruction image including a path according to the present invention, Figure 8 is a view for explaining the generation of a cross-sectional reconstruction image parallel or perpendicular to the path according to the present invention.

7 to 8, a multi-faceted reconstructed image including a path of a straight line or a curved line may generate an image by generating a line of scan lines in a predetermined direction set by the user at each point on the path according to the user's mouse operation. Can be. That is, when displaying a cross-sectional image on a three-dimensional image, it is possible to generate and display a reconstructed image if not only an image cut by a straight line but also a cut through various types of three-dimensional curves. The path reconstruction image formed at this time displays information on the surrounding structures around the path passing through the center line of the structure, and thus can be very useful for medical personnel when diagnosing an endoscope.

According to the present invention, in synthesizing a multi-dimensional reconstructed image using a virtual endoscope path, by performing curved planar reformation in a direction of observing a 3D image, the full length of the anatomical structure to be viewed can be shown. In addition, the curved reconstructed image may be rotated substantially based on the virtual endoscope path by changing the curved reconstructed image according to the rotation of the 3D image.

In addition, as shown in FIG. 8, in the case of the cross-sectional reconstruction of the linear interface method parallel to the virtual endoscope path from the virtual endoscope perspective, the direction of the virtual endoscope path can also be freely rotated about the axis, and the virtual endoscope path from the virtual endoscope viewpoint. The cross-sectional reconstruction of the straight interface method perpendicular to may be implemented to continuously show the vertical cross section of the structure to be observed while moving as the virtual endoscope moves.

That is, the virtual endoscope image findings can be easily understood through the reconstructed image by allowing the user to select and use a proper reconstruction method according to the characteristics of the structure to be observed from the virtual endoscope perspective.

First, the method of unfolding the full length of the anatomical structure to be seen by performing curved planar reformation in the direction of observing the 3D image is implemented as follows.

That is, the information on the curved path along the center of the structure is stored as, for example, three-dimensional coordinate values (x, y, z) as points separated by one length. Therefore, since we know a point on the path and a viewing vector, we get a surface reconstruction image along a straight line passing through a point on the path.

Starting at each point, we move forward with the direction vector until we get out of the boundary of the volume data, and we get the value corresponding to the pixel. Performing on one point yields a line of the reconstructed image.

Therefore, if all three-dimensional points are performed, one complete reconstruction image is obtained. 7 shows the process.

On the other hand, as shown in Figure 8, in order to reconstruct the cross-section or parallel to the path using a normal vector at a point on the path. Each point in the path has a normal vector value that is the derivative of the curve equation. Thus, in order to reconstruct a plane parallel to a point on the path, a planar reconstruction image is made of a plane containing one point on the path and one point on the normal vector. In this case, since several planes may come out, it may be desirable to select a plane by a user's manipulation.

Also, one plane is determined because the cross section perpendicular to one point of the path is a plane perpendicular to one point and the normal vector. Therefore, by selecting a point on the path, a planar reconstruction image can be made into a plane in which the normal vector at that point becomes the normal vector of the plane.

Next, a method of obtaining a 3D image effect on a 2D image, which is another feature of the present invention, will be described in detail.

9 is a view for explaining a three-dimensional camera adjustment in a two-dimensional image according to the present invention, in particular, a view for explaining the direction operation on the plane and the direction operation on the depth, respectively.

Referring to FIG. 9, a mouse click is used to designate a position to be observed in a 3D image made of a 2D image or a volume rendering image, and the virtual endoscope visual direction is a planar direction of x and y coordinates in the 2D image and the 3D image, respectively. You can also adjust the depth to the and z coordinates.

That is, according to the present invention, the virtual endoscope image is newly rendered whenever the viewpoint and the visual direction are changed, so that the user can easily select the position and position of the virtual endoscope to obtain a three-dimensional image effect in the two-dimensional image. .

The operation method is described below from the user's point of view using the virtual endoscope.

The position and orientation of the current camera is displayed in red in the volume rendering viewer and the two-dimensional section image. At this time, the left mouse is clicked to change the position of the camera, for example, and the right mouse is moved to change the direction. However, when a general two-dimensional image is used as a reference image of a virtual endoscope, it is difficult to display the depth direction in two dimensions.

However, in the virtual endoscope system according to the present invention, as shown in FIG. 8, the direction mode of the z coordinate can be changed. In other words, each time you right-click the arrow, the toggle mode changes. That is, red in the plane direction and blue in the depth direction can be adjusted. In red, the direction changes in a plane perpendicular to the direction of the camera as the mouse moves. In blue, the direction changes in the plane including the direction vector. It is possible.

FIG. 10 illustrates a 3D volume rendering image used as a reference image of the virtual endoscope system according to the present invention. In particular, FIG. 10 is an example showing an image of the bronchus visualized by 3D volume rendering for endoscopy.

Referring to FIG. 10, in order to view the bronchus with a virtual endoscope, the 3D volume rendering is first classified into density values corresponding to the bronchus, and unnecessary parts are deleted to show the bronchus and the direction of the camera or the user is adjusted. .

FIG. 11 illustrates an example of a screen provided by a preferred virtual endoscope system according to the present invention based on the 3D volume rendering image of FIG. 10.

Referring to FIG. 11, when entering the virtual endoscope system, four viewers as shown in FIG. 2 are shown, and the position of the camera is indicated by a red point in each viewer, and a direct path in the 3D reference image, which is one of the features of the present invention. Shows an example of setting.

When the path is set by the user's manipulation, the reconstructed image is generated if the path is included in the MPR path viewer. If the path is included, the position of the camera is also displayed in the reconstructed image. In order to change the direction of the reconstructed image including the path, a mouse is used in the volume rendering image viewer. In this case, the direction of the reconstructed image is configured to be displayed in a blue line in the three-dimensional volume rendering viewer among the four viewers.

FIG. 12 illustrates a screen providing a function of allowing a user to arbitrarily modify a path designated by the user in FIG. 11 in a preferred virtual endoscope system according to the present invention.

Referring to FIG. 12, a process of directly modifying a path in a 3D reference image is illustrated. To modify the path, use the mouse to move the point shown on the path.

In addition, the system according to the present invention can include a tool for measuring the length on the path thus made for the convenience of the user.

The virtual endoscope system according to the present invention, the virtual endoscope display method using the same, and a computer-readable recording medium storing a program for executing the method on a computer can be modified and applied in various forms within the scope of the technical idea of the present invention. It is not limited to the said preferable embodiment.

For example, the input unit may use not only a mouse but also a light pen, a keyboard, or various input devices. In addition to a system for medical image processing, the input unit may be used in the field of design and construction of a building having a three-dimensional structure such as an automobile, a ship, and a building. A wide range of applications is possible.

As described above, the virtual endoscope system according to the present invention uses an image generated by volume rendering without a preprocessing process for 3D volume data as a reference image, and the user easily sets a path using the reference image. It is possible to easily understand the surroundings of the structure by generating MPR images according to the set path.

In addition, the virtual endoscope described above is a combination of high technology as an integrated environment of various types of medical images, the virtual endoscope will be able to develop into an integrated system including more various images and various technologies due to the present invention.

Although the above has been described with reference to a preferred embodiment of the present invention, those skilled in the art will be able to variously modify and change the present invention without departing from the spirit and scope of the invention as set forth in the claims below. It will be appreciated.

As described above, the virtual endoscope according to the present invention is non-invasive to the patient, and allows the internal view of the organ without pain.

In addition, in an endoscope, the inside of the organ may be observed only in the direction in which the camera enters, that is, in the forward direction of the camera, but the virtual endoscope according to the present invention may be observed in the opposite direction in which the camera enters, that is, in the backward direction of the camera. Very useful.

In addition, until now, the virtual endoscope system has been difficult to operate due to the inefficiency of the interface. However, according to the present invention, the user can easily operate the camera and show the various reference images together. It can be increased further.

Claims (9)

delete An input / storing unit configured to receive and store volume data storing density values of three-dimensional structures having predetermined characteristics; An input unit configured to receive a path designated by a user on a 3D reference image; A display unit displaying the 3D reference image and displaying an endoscope image along a path; And A reference image processor which generates a 3D reference image displayed on the display unit from the volume data stored in the input / storage unit, and processes a path input through the input unit as curve data on the 3D reference image; Determine a position in the volume data of the voxel corresponding to the wall of the internal structure from which the endoscope image is to be viewed from the path processed by the reference image processor, and in the volume data of the voxel so that the path actually passes through the internal structure. A path converting unit correcting a path by a position; And an endoscope image processor having a multi-faceted reconstructed image synthesizer configured to generate a multi-faceted reconstructed image including the path according to the corrected path obtained by the path converter. The position in the volume data of the voxel corresponding to the wall of the inner structure corresponds to a first stored position and a wall of the inner structure which stores the position where a predetermined fired ray first meets the voxel corresponding to the wall of the inner structure. And a midpoint point of the internal structure determined based on a second stored location storing a secondary encounter with the voxel. In the virtual endoscope display method for storing the raw image data, classifying the density value of the stored raw image data to display a virtual endoscope, (a) generating a volume rendering image corresponding to an area to be observed from the 3D volume data; (b) displaying each of the generated volume rendering image, two-dimensional image, virtual endoscope image, and path reconstruction image on a plurality of viewers partitioning one screen; (c) updating the display of the path reconstruction image as a path is set in one of the 2D image and the 3D volume rendering image according to a user's manipulation; (d) checking whether to change the position of the camera according to a user's operation; And (e) When the camera position change request is made, generate virtual endoscope image data adjusted according to the position change of the camera, and update the virtual endoscope image and the 2D image based on the generated virtual endoscope image data. Steps to Virtual endoscope display method comprising a. The method of claim 3, and (f) checking whether or not the volume rendering image is requested to be remanufactured according to a user's operation, and feeding back to the step (a) if there is a remanipulation request. The method of claim 3, wherein step (c) comprises: (c-1) generating a 3D curve equation by adding a curved control point as a point is designated in a 2D image or a 3D volume rendered image according to a user's manipulation; (c-2) generating a curve path using the equation of the three-dimensional curve; And (c-3) storing the corresponding points at predetermined intervals along the generated curved path. The method of claim 5, wherein step (c-2) comprises: And using the 3D volume rendering image as a reference image, converting the image to the center of the structure. The method of claim 6, wherein converting to the center of the structure, Setting a minimum value and a maximum value of a depth to be inspected according to a mouse input in a 2D image drawn by volume rendering; Multiplying the temporal inverse to convert points in the screen coordinate system into points in the spatial coordinate system; Firing a beam of light at the set minimum depth; First storing the location when the emitted light beam first meets a voxel corresponding to a wall of an object; If the emitted light beam meets a voxel corresponding to a wall of an object for a second time, storing the second location; And And generating a midpoint result based on the first stored position and the second stored position. The method of claim 3, wherein the camera, The volume rendering image, the two-dimensional image, the virtual endoscope image, and the path reconstruction image are displayed by using a predetermined visual vector, and the volume rendering image, the two-dimensional image, and the virtual endoscope by using the changed time vector having adjusted the visual vector And a three-dimensional image effect on a two-dimensional image by updating the image and the path reconstruction image. A computer-readable recording medium storing a program for storing a raw image data, classifying density values of stored raw image data, and displaying a virtual endoscope on the computer. (a) generating a volume rendering image corresponding to an area to be observed from the 3D volume data; (b) displaying each of the generated volume rendering image, two-dimensional image, virtual endoscope image, and path reconstruction image on a plurality of viewers partitioning one screen; (c) updating the display of the path reconstruction image as a path is set in one of the 2D image and the 3D volume rendering image according to a user's manipulation; (d) checking whether to change the position of the camera according to a user's operation; And (e) When the camera position change request is made, generate virtual endoscope image data adjusted according to the position change of the camera, and update the virtual endoscope image and the 2D image based on the generated virtual endoscope image data. Steps to A computer-readable recording medium storing a program for executing a program on a computer.