JP2013094438A - Image processing device, method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To display positional relationships between an attention point set on a tomographic image and a prescribed structure in such a way as clearly recognizable when the tomographic image representing a part of the prescribed structure such as a blood vessel is displayed on a screen for observation.SOLUTION: The image processing device includes: a structure extracting means 10 for extracting a tree structure Ms representing the prescribed structure extending widely while repeatedly branching from one point of origin from a three-dimensional image; an attention point setting means 20 for setting an attention point S on the tree structure Ms according to the input by an operator specifying an optionally selected position on the three-dimensional image; a projected image generating means 30 for generating a projected image Ij' by superimposing an optionally selected tomographic image Ij including the attention point S on a projected image obtained by projecting a partial structure Pj constituting a passage from the attention point S to a point 0 representing the point of origin thereof among the whole tree structure Ms on the tomographic image Ij; and a display means 40 for displaying the generated projected image.

Description

  The present invention provides an image processing apparatus and method for easily displaying the positional relationship between a target point and the predetermined structure when the target point is set on an image representing a predetermined structure such as a blood vessel. And program.

  Conventionally, when examining the state of a blood vessel using tomographic images such as CT, MRI, etc., in which blood vessels are imaged, an operator such as a doctor or the like has to make continuous tomographic images in order to confirm whether the portion of interest is an artery or a vein. The process of switching and displaying the images sequentially and tracing the blood vessel from the target part to the root is performed manually. However, this method places a heavy burden on the operator, so a more efficient confirmation method is desired. Yes.

  On the other hand, in Patent Document 1, when displaying a tomographic image in which a point of interest such as a lesion is displayed, a projected image obtained by projecting a bone such as a rib or a vertebra is displayed in an overlapping manner, thereby determining the position of the lesion. A method has been proposed that makes it easy to grasp the relationship with the bone position. Therefore, when this method is applied and the tomographic image of the blood vessel is displayed, the projected image obtained by projecting the structure of the entire blood vessel is displayed in an overlapping manner, so that a portion of interest can be obtained only by observing one display image. It is conceivable to make it possible to grasp the position of the entire blood vessel including that part of the blood vessel.

JP 2008-93254 A

  However, in the above-described method of superimposing and displaying a projection image that projects the entire structure of the blood vessel, the longer the total length of the extracted blood vessel or the more complicated the structure of the blood vessel, the more the projected image of the blood vessel that is superimposed and displayed. However, since it becomes an obstacle to the observation of the part of interest, there is a problem that the observation performance deteriorates.

  In view of the above circumstances, the present invention makes it easy to recognize the positional relationship between a point of interest set on an image and the predetermined structure when an image showing the predetermined structure is displayed on the screen and observed. An object of the present invention is to provide an image processing apparatus, a method, and a program for displaying.

  An image processing apparatus according to the present invention includes a structure extraction unit that extracts a tree structure representing a predetermined structure extending and expanding from one start portion while repeating branching, and an arbitrary position on the three-dimensional image. In response to an input by the operator who designates the attention point, the attention point setting means for setting the attention point on the extracted tree structure, an arbitrary tomographic image including the set attention point, and the extracted tomographic image A projection image generating means for generating a projection image formed by superimposing a projection image obtained by projecting a partial structure constituting a path from a point of interest to a point representing a starting portion of the whole tree structure, and the generated And display means for displaying a projected image.

  Here, the tree structure refers to a graph having a structure that extends and expands while repeating branching from one node among graphs composed of a node group and an edge group representing a connection relationship between the nodes.

  In the image processing apparatus, the structure extraction unit extracts a plurality of tree structures representing two or more predetermined structures extending and expanding while repeating branching from one starting part, respectively, from a three-dimensional image. The attention point setting means sets an attention point at any position on the extracted tree structure based on the position on the designated three-dimensional image, and the projection image generation means comprises: When a point of interest is set on the downstream side of the connected portion on the plurality of tree structures having portions connected to each other by the point of interest setting means or the connected portion, any fault including the set point of interest An image and a partial structure constituting a path from the point of interest to a point representing each starting portion of the plurality of tree structures having the connected portion in the tomographic image in the whole of the plurality of tree structures extracted projection It may be configured to generate a projection image composed by superimposing the projection images were.

  Here, the “downstream side” means a side away from the starting portion.

  Further, in the image processing apparatus, the structure extraction unit extracts a plurality of tree structures representing two or more predetermined structures extending and expanding while repeating branching from one starting part, respectively, from a three-dimensional image. Yes, the attention point setting means sets the attention point at any position on the extracted plurality of tree structures based on the position on the designated three-dimensional image. If the attention point is set on the downstream side of the adjacent part on the plurality of tree structures having the parts adjacent to each other by the attention point setting means, the arbitrary points including the set attention point And a tomographic image of the tree structure, a point of interest on the tree structure in which the point of interest is set among all of the plurality of tree structures extracted in the tomographic image, and a point representing the starting portion thereof and the adjacent part s tree It may be configured to generate a projection image composed by superimposing the projection images obtained by projecting the partial structure constituting the path from the adjacent portion of the concrete to the point representing the proximal portion.

  Here, whether or not a part on the tree structure is a part close to another tree structure is determined by, for example, whether or not the minimum value of the interval between the part and the other tree structure is equal to or less than a predetermined threshold value. be able to. At this time, the threshold value can be set based on the radius of the predetermined structure.

  Also, whether or not an edge constituting a part of a tree structure is a part adjacent to another tree structure is, for example, whether there are two or more pairs of edges that can be connected so as to form a straight line in the vicinity of the edge. It can be determined by whether or not. This is because when there are two or more pairs of edges that can be connected to form a straight line, there is a high possibility that two or more straight lines connecting the pairs exist in the vicinity of the edge. At this time, the vicinity range can be set based on the radius of a predetermined structure.

  In addition, the above-mentioned “from a close part” means that, when the close part covers a certain range, from any position within the range.

  Further, in the image processing apparatus, the projection image generation unit is configured to output the designated second tomographic image and the second tomographic image in response to an input from an operator who designates an arbitrary tomographic image that intersects the partial structure. The image is formed by superimposing a projection image obtained by projecting the second partial structure constituting the path from the point where the second tomographic image of the entire partial structure intersects the partial structure to the point representing the starting part. The display device may further include a function of generating a second projection image, and the display unit may switch the generated second projection image to the projection image or display it together with the projection image.

  Further, when the projection image generating means generates the projection image by projecting two or more partial structures of different tree structures, the display colors of the projection images may be made different from each other so that they can be distinguished from each other. Good.

  The image processing method of the present invention is a method in which the processing performed by each unit of the image processing apparatus is executed by at least one computer.

  An image processing program of the present invention is a program that causes at least one computer to execute the image processing method. This program is recorded on a recording medium such as a CD-ROM or DVD, or recorded in a state where it can be downloaded to a storage attached to a server computer or a network storage, and provided to the user.

  According to the image processing device, method, and program of the present invention, a tree structure representing a predetermined structure extending and extending while repeating branching from one starting portion is extracted from a three-dimensional image, and an arbitrary one on the three-dimensional image is extracted. In response to an input by the operator who designates the position of the image, a point of interest is set on the extracted tree structure, an arbitrary tomographic image including the set point of interest, and the tree extracted in the tomographic image A projection image is generated by superimposing a projection image that projects a partial structure that constitutes the path from the point of interest to the point that represents the beginning of the entire structure, and the generated projection image is displayed. Therefore, the positional relationship between the attention point and the predetermined structure can be easily displayed.

  For example, when examining the state of a blood vessel using a tomographic image, it is sufficient to confirm whether the part where the point of interest is set is an artery or a vein, from that part to the origin of the blood vessel. By superimposing and displaying only a limited portion used for the confirmation, an image suitable for observation can be provided as compared with the method of superimposing and displaying the entire blood vessel.

  In the image processing apparatus, method and program of the present invention, a plurality of tree structures each representing two or more predetermined structures extending and extending while repeating branching from one starting portion are extracted from a three-dimensional image, Based on the position on the specified three-dimensional image, an attention point is set at any position on the extracted plurality of tree structures, and the connection points on the plurality of tree structures having portions connected to each other there are connected therewith. When a point of interest is set downstream of a portion or a connected portion thereof, an arbitrary tomographic image including the set point of interest and a whole of the plurality of tree structures extracted in the tomographic image When a projection image is generated by superimposing a projection image formed by projecting a partial structure that forms a path from a point of interest to a point that represents the starting portion of each of the plurality of tree structures having the connected portions. Can recognize the positional relationship between a point of interest and all of those predetermined structures when it is unclear which of the two or more predetermined structures is the point of interest. It is possible to prevent the occurrence of a situation in which an erroneous image analysis is performed by displaying only the positional relationship with one of the erroneous structures.

  In addition, a plurality of tree structures representing two or more predetermined structures extending and extending while repeating branching from one starting portion are extracted from the three-dimensional image, and based on the designated positions on the three-dimensional image. Then, a point of interest is set at any position on the extracted plurality of tree structures, and the adjacent portions on the plurality of tree structures having portions close to each other there or on the downstream side of the adjacent portions. When an attention point is set, an arbitrary tomographic image including the set attention point and an attention on the tree structure in which the attention point is set out of all of the extracted tree structures in the tomographic image A projected image obtained by projecting a partial structure constituting a path from a point to a point representing the starting part and from the adjacent part on the other tree structure having the neighboring part to the point representing the starting part. Superimposed projection image If it is generated, if multiple tree structures have parts close to each other, there may be an error in extracting the tree structure due to the intersection of two or more predetermined structures in that part. From the above, although the point of interest is initially recognized as a point on one predetermined structure, it is unclear which of the two or more predetermined structures is the point of interest. In such a case, the positional relationship between the attention points and all of those predetermined structures can be displayed in a recognizable manner, and incorrect image analysis can be performed by displaying only the positional relationship with one of the wrong structures. Can be prevented from occurring.

1 is a diagram illustrating a schematic configuration of an image processing apparatus according to a first embodiment. The figure which shows an example of the tree structure extracted by the structure extraction means The figure for demonstrating the production | generation process of the projection image by a projection image production | generation means The figure for demonstrating the production | generation process of the 2nd projection image by a projection image production | generation means The figure which shows schematic structure of the image processing apparatus which concerns on 2nd Embodiment. The figure which shows an example of the tree structure extracted by the structure extraction means (the 1) The figure for demonstrating the production | generation process of the projection image by a projection image production | generation means (the 1) The figure which shows an example of the tree structure extracted by the structure extraction means (the 2) The figure for demonstrating the production | generation process of the projection image by a projection image production | generation means (the 2) The figure for demonstrating the determination method whether a specific edge is a part which adjoins other tree structures

  Embodiments of an image processing apparatus and method and a program according to the present invention will be described below. FIG. 1 is a schematic configuration diagram of an image processing apparatus 1 according to an embodiment of the present invention. The configuration of the image processing apparatus 1 as shown in FIG. 1 is realized by executing an image processing program read into the auxiliary storage device on the computer. At this time, the image processing program is stored in a storage medium such as a CD-ROM or distributed via a network such as the Internet and installed in a computer. The image processing apparatus 1 in FIG. 1 provides a display more suitable for observation when a doctor or the like displays a tomographic image such as CT or MRI on a screen and observes the state of a blood vessel. Means 10, attention point setting means 20, projection image generation means 30, and display means 40 are provided. Hereinafter, in the present embodiment, a case where the predetermined structure in the present invention is a hepatic vein will be described.

  The structure extraction means 10 extracts a tree structure representing the hepatic vein from the three-dimensional image V representing the liver. Specifically, first, a region having an image feature and / or a structural feature of the hepatic vein is detected from the three-dimensional image V, and the thin line obtained by thinning the region is converted into a branch point and a predetermined distance. A graph is generated by dividing. Here, the three-dimensional image V is a number of continuous tomographic images Ii (i = 1 to n: n are tomographic images) acquired by, for example, a CT apparatus, an MRI apparatus, an ultrasonic diagnostic apparatus or the like stored in the data storage means. Number).

  Next, among a plurality of candidate points (nodes) constituting the generated graph, a plurality of candidate points forming a structure that matches or is most similar to the hepatic vein shape model Mref stored in the shape model storage means DB Is selected as a corresponding point corresponding to each of a plurality of teacher labels (nodes) constituting the shape model Mref, and the selected plurality of corresponding points are connected so as to substantially match the shape model Mref, thereby the tree structure Ms. Is generated and extracted. Thereby, for example, a tree structure Ms as shown in FIG. 2 is extracted. Here, the shape model Mref is a tree-structured graph representing the general shape of the hepatic vein, and can be obtained by learning a large number of sample images, for example.

The corresponding points are selected by, for example, obtaining an optimal solution (corresponding relationship) that achieves minimization of the cost function F expressed by the following equation (1). Here, R represents a set of correspondences between candidate points and teacher labels in the allowable solution x, θa is a cost for an arbitrary correspondence a belonging to the set R, and θab is two correspondences a belonging to the set R. And the cost for the combination of b (see Japanese Patent Application Laid-Open No. 2011-98195).

  The attention point setting means 20 sets the attention point S on the extracted tree structure Ms in response to an input by an operator who designates an arbitrary position on the three-dimensional image V. Specifically, for a tomographic image Ij (jεi) displayed on an image display device such as a display, the input device 2 such as a mouse or a keyboard that designates an arbitrary position on the image display device. In response to an input from the operator, a position on the tomographic image Ij corresponding to the designated position is acquired, and a position on the three-dimensional image V corresponding to the acquired position on the tomographic image Ij is specified. . When the specified position on the three-dimensional image V is on the tree structure Ms, the tree structure closest to the specified position is at that position, and when the specified position is not on the tree structure Ms. An attention point S is set at the upper position.

  The projection image generating means 30 projects a tomographic image including the set attention point S and a partial structure constituting a path from the attention point to a point representing the starting portion of the entire tree structure Ms on the tomographic image. A projection image is generated by superimposing the projected image. Specifically, first, as shown in the left diagram of FIG. 3, the partial structure Pj (indicated by a bold line in the figure) that configures the path from the point of interest S to the point O that represents the starting part of the tree structure Ms is specified. Then, a projection image Pj ′ obtained by projecting the identified partial structure Pj onto the tomographic image Ij is generated. Then, a projection image Ij ′ is generated by superimposing the generated projection image Pj ′ and the tomographic image Ij. Thereby, for example, a projection image Ij ′ as shown in the right diagram of FIG. 3 is generated. The projection image Ij ′ generated by the projection image generation unit 30 is displayed on the image display device 3 such as a display by the display unit 40.

  In addition, after the projection image Ij ′ is generated and displayed, the projection image generation unit 30 selects another tomographic image Ik (kεi, k ≠ j) located on the start side from the tomographic image Ij. When an operator input from the input device 2 such as a mouse or a keyboard instructing display on the image display device is received, a projection image Ik ′ is further generated. Specifically, as shown in the left diagram of FIG. 4, from a point S ′ where the tomographic image Ik of the whole partial structure Pj intersects the partial structure Pj to a point O representing the starting portion of the tree structure Ms. A second partial structure Pk (indicated by a bold line in the figure) that constitutes the path is identified, and a projection image Pk ′ obtained by projecting the identified second partial structure Pk on the tomographic image Ik is generated. Then, a second projection image Ik ′ is generated by superimposing the generated projection image Pk ′ and the tomographic image Ik. Thereby, for example, a second projection image Ik ′ as shown in the right diagram of FIG. 4 is generated. The generated second projection image Ik ′ is displayed on the image display device 3 by the display means 40 while switching to the projection image Ij ′ or together with the projection image Ij ′.

  Next, a second embodiment of the image processing apparatus and method and program of the present invention will be described. FIG. 5 is a schematic configuration diagram of an image processing apparatus 101 according to the second embodiment of the present invention. The configuration of the image processing apparatus 101 as shown in FIG. 5 is realized by executing an image processing program read into the auxiliary storage device on the computer. At this time, the image processing program is stored in a storage medium such as a CD-ROM or distributed via a network such as the Internet and installed in a computer. The image processing apparatus 101 in FIG. 5 provides a display more suitable for observation when a doctor or the like displays a tomographic image such as CT or MRI on the screen and observes the state of the blood vessel. Means 110, attention point setting means 120, projection image generation means 130, and display means 140 are provided. Hereinafter, in the present embodiment, a case will be described in which a plurality of predetermined structures in the present invention are two blood vessels of a portal vein and a hepatic artery in the liver.

  The structure extraction unit 110 extracts the tree structure Ms1 representing the portal vein and the tree structure Ms2 representing the hepatic artery from the three-dimensional image V representing the liver. Specifically, first, a region having an image characteristic and / or structural feature of a blood vessel is detected from the three-dimensional image V, and the thin line obtained by thinning the region is divided by a branch point and a predetermined distance. To generate a graph. A structure that matches or most closely resembles the portal vein shape model Mref1 and hepatic artery shape model Mref2 stored in the shape model storage means DB from among a plurality of candidate points (nodes) constituting the generated graph. A plurality of candidate points to be formed are selected as corresponding points corresponding to each of the plurality of teacher labels constituting the shape model Mref1 and the shape model Mref2, and the selected plurality of corresponding points are abbreviated as the shape model Mref1 and the shape model Mref2. The portal tree structure Ms1 and the hepatic artery tree structure Ms2 are generated and extracted by connecting them so as to match. Thereby, for example, a tree structure Ms1 and a tree structure Ms2 as shown in FIGS.

  Here, the shape model Mref1 is a graph of a tree structure representing the general shape of the portal vein, and can be obtained by learning a large number of sample images representing the portal vein. It is a graph of a tree structure representing a general shape of an artery, and can be obtained by learning a large number of sample images representing the hepatic artery.

  The point-of-interest setting means 120 sets the point of interest at any position on the extracted tree structure in response to an input by an operator who designates an arbitrary position on the three-dimensional image V. Specifically, for a certain tomographic image Ii displayed on an image display device such as a display, the operator inputs using an input means such as a mouse or a keyboard for designating an arbitrary position on the image display device. Accordingly, a position on the tomographic image Ii corresponding to the designated position is acquired, and a position on the three-dimensional image V corresponding to the acquired position on the tomographic image Ii is specified. When the specified position on the three-dimensional image V is on the tree structure Ms1 or the tree structure Ms2, the specified position is not at any position on the tree structure Ms1 or the tree structure Ms2. In this case, the attention point S is set at any position on the tree structure Ms1 and the tree structure Ms2 closest to the specified position.

  The projection image generation unit 130 superimposes the tomographic image including the set attention point S and the projection image obtained by projecting a part of the extracted tree structure Ms1 and tree structure Ms2 on the tomographic image. A projection image is generated. The projection image generation method differs depending on the positional relationship between the tree structure Ms1 and the tree structure Ms2 and the set position of the attention point S.

  As shown in FIG. 6, when the tree structure Ms1 and the tree structure Ms2 have a portion J where the tree structures Ms2 are connected to each other, and the attention point S is set downstream of the connected portions, the projection image generation means 130 As shown in the left diagram of FIG. 7, both paths from the point of interest S to a point O1 representing the start of the tree structure Ms1 and from the point of interest S to a point O2 representing the start of the tree structure Ms1 are configured. A partial structure Pj (indicated by a bold line in the figure) is specified, and a projection image Pj ′ is generated by projecting the specified partial structure Pj on the tomographic image Ij. Then, a projection image Ij ′ is generated by superimposing the generated projection image Pj ′ and the tomographic image Ij. Thereby, for example, a projection image Ij ′ as shown in the right diagram of FIG. 7 is generated.

  On the other hand, as shown in FIG. 8, when the tree structure Ms1 and the tree structure Ms2 have a portion J that is close to each other, and the attention point S is set on the downstream side of the close portion, the projection image generation means 130 is provided. As shown in FIG. 9, the proximity on the other tree structure Ms2 having the adjacent portion from the attention point S on the tree structure Ms1 to which the attention point S is set to the point O1 representing the starting portion thereof. A partial structure Pj (indicated by a bold line in the figure) that constitutes a path from the portion to the point representing the starting portion O2 is specified, and a projection image Pj ′ is generated by projecting the specified partial structure Pj on the tomographic image Ij To do. Then, a projection image Ij ′ is generated by superimposing the generated projection image Pj ′ and the tomographic image Ij.

  At this time, as shown in the left diagram of FIG. 8, the point of interest S is determined as to whether or not it is a portion adjacent to the other tree structure Ms2 upstream of the point of interest S of the tree structure Ms1 in which the point of interest S is set. The edge on the tree structure Ms1 to which is set is the edge E0, and a plurality of edges constituting the shortest path from the edge E0 to the point O1 representing the starting portion of the tree structure Ms1 are edge Ei (i = 1 to n: n). Is the number of edges constituting the shortest path), and each of the edge E0 and the edge Ei is examined by sequentially determining whether or not the edge is a part close to the other tree structure Ms2.

  Here, whether or not a specific edge is close to another tree structure is determined by examining whether or not there are two or more pairs of edges that can be connected so as to form a straight line near each other. judge. Here, whether or not a pair of edges can be connected to form a straight line depends on whether or not the shortest distance between the straight lines including each edge and the angle between the edges are each equal to or less than a predetermined threshold value. (See Japanese Patent Application No. 2011-060193). For example, as shown in FIG. 10, when there are five edges Ea, Eb, Ec, Ed, Ee in the vicinity of the edge Ei to be determined, a pair of edge Ea and edge Ed, a pair of edge Eb and edge Ee Can be determined as a pair of edges that can be connected to form a straight line.

  Further, when the projection image generating means 130 generates the projection image Ij ′ by each of the above methods, the projection is generated for each tree structure such that the portal tree structure Ms1 is blue and the hepatic artery tree structure Ms2 is red. The display colors of the images are differentiated from each other. The projection image Ij ′ generated in this way is displayed on the image display device 3 such as a display by the display means 140.

  The projection image generation unit 130 further generates another tomographic image Ik (kεi, k ≠ j) located on the start side of the tomographic image Ij after the projection image Ij ′ is generated and displayed. When an operator input from the input device 2 such as a mouse or a keyboard instructing display on the image display device is received, a second projection image Ik ′ is further generated. Specifically, the second partial structure Pk constituting a path from the point S ′ where the tomographic image Ik of the whole partial structure Pj intersects the partial structure Pj to the point representing the starting part of the tree structure. And the projection image Pk ′ obtained by projecting the specified second partial structure Pk onto the tomographic image Ik is generated. Then, a second projection image Ik ′ is generated by superimposing the generated projection image Pk ′ and the tomographic image Ik. The generated second projection image Ik ′ is displayed on the image display device 3 by the display unit 140 while switching to the projection image Ij ′ or together with the projection image Ij ′.

  As described above, according to each of the above-described embodiments, the structure extraction means 10; 110 represents the tree structure Ms; Ms1, Ms2 representing a blood vessel extending and expanding while repeating branching from one starting portion. Extracted from the image V, the attention point setting means 20; 120 sets the attention point S on the extracted tree structure in response to an input by an operator who designates an arbitrary position on the three-dimensional image V. Projection image generation means 30; 130 from an arbitrary tomographic image Ij including the set attention point S to the point representing the starting portion of the whole tree structure extracted in the tomographic image Ij. Since the projection image Ij ′ is generated by superimposing the projection image Pj ′ on which the partial structure Pj that constitutes the path is superimposed, and the display means 40; 140 displays the generated projection image Ij ′. Attention point S for operators such as doctors It can be displayed easily recognize the positional relationship between the origin of the tube. By this display, a skilled doctor or the like can easily recognize whether the portion of the blood vessel where the attention point S is set is an artery or a vein.

  Further, according to the second embodiment, when it is unclear which of the two blood vessels of the portal vein and hepatic artery the set attention point S is, Since the positional relationship between S and all those blood vessels is displayed in a recognizable manner, an erroneous image analysis may be performed by displaying only the positional relationship with one of the erroneous structures. The occurrence of a situation can be prevented.

  In each of the above embodiments, the case where the predetermined structure is a liver blood vessel such as a hepatic vein or portal vein is illustrated, but the predetermined structure repeats branching from one starting portion. However, it may be a structure that extends and extends, and may be, for example, a lung or a blood vessel of the heart.

  In each of the above embodiments, the case where the attention point setting means 20; 120 sets the attention point S according to the input of the operator with respect to the display of the tomographic image Ij is exemplified. You may set the attention point S according to the input of the operator with respect to the display of the image V (for example, display of a volume rendering image).

  In the second embodiment, the projection image generation unit 130 determines whether or not there is a portion close to another tree structure upstream of the attention point S of the tree structure in which the attention point S is set. In the above description, the upstream edge is determined by examining whether or not there are two or more edge pairs that can be connected so as to form a straight line in the vicinity thereof. The determination may be made by examining whether or not there is a portion where the minimum value of the interval between the portion and another tree structure is a predetermined threshold value or less.

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 10 Structure extraction means 20 Attention point setting means 30 Projection image generation means 40 Display means DB Shape model memory | storage means Ii Tomographic image Ii 'Projection image Mref Shape model Ms Tree structure O Point Pj which represents the beginning part Partial structure Pj ′ Projected image S Attention point V 3D image

Claims (10)

A structure extraction means for extracting from a three-dimensional image a tree structure representing a predetermined structure extending and extending while repeating branching from one starting portion;
Attention point setting means for setting an attention point on the extracted tree structure in response to an input by an operator who designates an arbitrary position on the three-dimensional image;
An arbitrary tomographic image including the set attention point and a partial structure constituting a path from the attention point to a point representing the starting portion of the whole extracted tree structure are projected onto the tomographic image. Projection image generation means for generating a projection image formed by superimposing the projection image;
An image processing apparatus comprising: display means for displaying the generated projection image. The structure extraction means extracts a plurality of tree structures representing two or more predetermined structures extending and extending while repeating branching from one starting part, respectively, from the three-dimensional image,
The attention point setting means sets the attention point at any position on the extracted tree structure based on the designated position on the three-dimensional image;
When the projected image generation means sets the attention point on the downstream side of the part on the plurality of tree structures having parts connected to each other by the attention point setting means, the set attention point And a path from the point of interest to a point representing the starting portion of each of the plurality of tree structures having the portion in the entire tree structure extracted in the tomographic image The image processing apparatus according to claim 1, wherein the projection image is generated by superimposing a projection image on which the partial structure is projected. The structure extraction means extracts a plurality of tree structures representing two or more predetermined structures extending and extending while repeating branching from one starting part, respectively, from the three-dimensional image,
The attention point setting means sets the attention point at any position on the extracted tree structure based on the designated position on the three-dimensional image;
When the projected image generating means sets the attention point on the downstream side of the adjacent portion on the plurality of tree structures having portions adjacent to each other or the adjacent portion by the attention point setting means, the setting An arbitrary tomographic image including the focused point of interest, and a point representing the starting portion from the focused point on the tree structure in which the focused point is set among the plurality of tree structures extracted in the tomographic image And the projected image formed by superimposing the projected image formed by projecting the partial structure constituting the path from the adjacent portion on the other tree structure having the adjacent portion to the point representing the starting portion. The image processing apparatus according to claim 1, wherein the image processing apparatus is generated. The projection image generation means is designated in the designated tomographic image and the whole of the partial structure in the tomographic image in response to an input from an operator who designates an arbitrary tomographic image that intersects the partial structure. A second projection image formed by superimposing a projection image obtained by projecting a second partial structure constituting a path from a point where the tomographic image intersects the partial structure to a point representing the starting portion Further having
4. The image processing according to claim 1, wherein the display unit is configured to switch the generated second projection image to the projection image or display the projection image together with the projection image. 5. apparatus.   In the case where the projection image generation means generates the projection image by projecting two or more partial structures having different tree structures, the display colors of the projection images are different for each tree structure so as to be distinguishable from each other. The image processing apparatus according to any one of claims 2 to 4. A process of extracting from a three-dimensional image a tree structure representing a predetermined structure extending and extending while repeating branching from one starting portion;
Processing for setting a point of interest on the extracted tree structure in response to an input by an operator who designates an arbitrary position on the three-dimensional image;
An arbitrary tomographic image including the set attention point and a partial structure constituting a path from the attention point to a point representing the starting portion of the whole extracted tree structure are projected onto the tomographic image. Processing for generating a projection image formed by superimposing the projection image;
An image processing method, wherein the process of displaying the generated projection image is executed by at least one computer. The tree structure extraction process is to extract a plurality of tree structures representing two or more predetermined structures extending and expanding while repeating branching from one starting part, respectively, from the three-dimensional image,
The attention point setting processing is to set the attention point at any position on the extracted plurality of tree structures based on the designated position on the three-dimensional image.
In the projection image generation process, when the attention point is set on the downstream side of the part on the plurality of tree structures having parts connected to each other by the attention point setting process, the set Arbitrary tomographic image including a point of interest and a path from the point of interest to a point representing the starting part of each of the plurality of tree structures having the portion of the whole tree structure extracted in the tomographic image The image processing method according to claim 6, wherein the projection image is generated by superimposing a projection image obtained by projecting the partial structure constituting the image. The tree structure extraction process is to extract a plurality of tree structures representing two or more predetermined structures extending and expanding while repeating branching from one starting part, respectively, from the three-dimensional image,
The attention point setting processing is to set the attention point at any position on the extracted plurality of tree structures based on the designated position on the three-dimensional image.
When the attention point is set on the downstream side of the adjacent part on the plurality of tree structures having the part adjacent to each other by the setting process of the attention point or the adjacent part in the generation process of the attention point, An arbitrary tomographic image including the set attention point, and a starting portion from the attention point on the tree structure in which the attention point is set among all of the plurality of tree structures extracted in the tomographic image. The projection formed by superimposing a projection image obtained by projecting the partial structure that constitutes a path from the adjacent portion on the other tree structure having the adjacent portion to the point to be represented to the point representing the starting portion. The image processing method according to claim 6, wherein an image is generated. In the projection image generation process, the designation of the designated tomographic image and the whole of the partial structure in the tomographic image in response to an input from an operator who designates an arbitrary tomographic image that intersects the partial structure. A second projection image is generated by superimposing a projection image obtained by projecting a second partial structure constituting a path from a point where the obtained tomographic image intersects the partial structure to a point representing the starting portion. Is,
The image processing according to any one of claims 6 to 8, wherein the display processing is to display the generated second projection image by switching to the projection image or together with the projection image. Method. A structure extracting means for extracting from a three-dimensional image a tree structure representing a predetermined structure extending and extending from a starting portion of a computer while repeating branching;
Attention point setting means for setting an attention point on the extracted tree structure in response to an input by an operator who designates an arbitrary position on the three-dimensional image;
An arbitrary tomographic image including the set attention point and a partial structure constituting a path from the attention point to a point representing the starting portion of the whole extracted tree structure are projected onto the tomographic image. Projection image generation means for generating a projection image formed by superimposing the projection image;
The image processing program for functioning as a display means which displays the produced | generated projection image.