JP5745390B2 - Image processing apparatus, method, and program - Google Patents

Description

  The present invention relates to an image processing apparatus, method, and program for specifying the direction of blood flow in a partial blood vessel branched from a branch source branch in a blood vessel including a loop-shaped portion.

  When performing surgery on various organs of the human body, such as the liver and lungs, it is important to plan carefully how to perform the surgery before surgery in order to ensure that the surgery can proceed more smoothly. When planning an operation, it is necessary to accurately grasp the travel of blood vessels in the target organ and the direction of blood flow in each blood vessel.

  Patent Document 1 discloses a tree structure in which the direction of blood flow in a blood vessel represents the blood vessel in order to be able to appropriately determine the region to be removed when performing an operation to remove an affected part including a tumor in a specific organ. A region of a tree structure in which the target blood vessel is imaged is extracted from the image data based on the recognition that the direction is to move away from the starting portion (root), and the extracted region is an image display device. Based on the operator's input that specifies an arbitrary position on the displayed area, the control area from the specified position to the peripheral part can be identified and distinguished from other areas. It has been proposed to display.

Japanese Patent No. 4688361

  However, if the target blood vessel has a tree structure such as heart, liver, lung, kidney, stomach, etc., the path to follow the structure away from the origin is always unique. Although the direction of blood flow can be specified by the method described in Patent Document 1 in which the direction of blood flow is the direction of blood flow in the blood vessel, the target blood vessel has a loop shape as shown in FIG. In the case of a blood vessel including a portion, there are a plurality of paths that follow the structure to the side away from the starting portion, and the direction to follow is simply specified as the direction of blood flow. This method has a problem that the direction of blood flow cannot be correctly specified. FIG. 7 shows a blood vessel of the large intestine, and a loop-shaped portion exists in a portion A or the like surrounded by a dotted line in the drawing.

  In view of the above circumstances, the present invention provides an image processing apparatus, method, and program capable of specifying the direction of blood flow in a partial blood vessel branched from a branch source branch in a blood vessel including a loop-shaped portion. It is for the purpose.

  According to an image processing apparatus of the present invention, a reference point setting unit that sets a reference point on a graph structure representing a blood vessel structure including a loop-shaped portion acquired in advance according to an input by an operator via an input device; A downstream determination means for determining a downstream side of the blood flow from the reference point at a branch where the set reference point of the graph structure is located; and a downstream side of the determined blood flow from the reference point of the graph structure Arbitrary branch nodes located on the route to be traced have a direction based on the positions of a plurality of nodes located on the branch source branch and have a direction toward the branch node. A direction based on the first vector defined and the positions of a plurality of nodes located on any one branch branched from the branch source branch, and the direction away from the branch node A second vector defined to have Using the angle θ (0 <θ <180 °), the likelihood that the direction of blood flow in that branch is the direction away from the branch node is calculated so that the likelihood increases as the angle θ decreases. And determining means for determining whether the direction of blood flow in the one branch is a direction away from the branch node or a direction toward the branch node using the calculated likelihood. Features.

  Here, the graph structure refers to a structure composed of a node group and an edge group representing a connection relationship between the nodes.

  In the image processing apparatus of the present invention, the determination unit defines the likelihood based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to the plurality of nodes located on the branch source branch. A value that is larger as the first representative value is larger than a second representative value defined based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to the plurality of nodes located on the one branch. It may be calculated as follows. Here, the representative values (first representative value and second representative value) defined based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel are the diameters of the blood vessels at each of the plurality of positions on the blood vessel. It means a value that increases as the overall size of increases. Typical examples include an average value, a maximum value, and a minimum value.

In this case, the likelihood L can be calculated by the following equation (1), for example. In the following equation (1), V1 represents the first vector, V2 represents the second vector, and F (V1, V2) represents the smaller the angle θ formed by the first vector V1 and the second vector V2. Represents a function with a large value. R1 represents a first representative value, R2 represents a second representative value, and G (R1, R2) is a difference obtained by subtracting the second representative value R2 from the first representative value R1 (or the second representative value). This represents a function that increases as the ratio of the first representative value R1 to the representative value R2 increases. W1 and W2 are coefficients for weighting the above functions.

  In the image processing apparatus of the present invention, the determination unit does not use the first representative value or the second representative value, and based on only the information about the angle θ formed by the first vector V1 and the second vector V2. In the case of making a determination, the determination can be performed using the angle θ itself as the likelihood. For example, when the angle θ is greater than a predetermined threshold, the determination unit determines that the direction of blood flow in the one branch is a direction toward the branch node, and when the angle θ does not exceed the threshold, It can be determined that the direction of blood flow in the branch of the book is the direction away from the branch node.

  In the image processing apparatus according to the present invention, the determination unit performs determination using the likelihood for each of a plurality of branches branched from the branch source branch. As a result, the direction of blood flow in the plurality of branches is determined. If it is determined that the direction is all toward the branch node, the direction of blood flow in one branch having the highest likelihood calculated for the branch among the plurality of branches is a direction away from the branch node. It may be determined that there is.

  Further, in the image processing apparatus of the present invention, the downstream side determination unit is configured to select one of both sides of the reference point based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to the plurality of nodes positioned before and after the reference point. The average value of the diameter of the blood vessel may be determined as the downstream side of the blood flow, or the branch where the reference point of the graph structure is located is displayed on a display device such as a display, and the reference point It is also possible to receive an operator input from an input device such as a mouse or a keyboard that selects either one of the two sides and determine the selected one as the downstream side of the blood flow.

  Further, the image processing apparatus of the present invention sets the graph structure in the entire region representing the blood vessel in the image representing the blood vessel in response to an input by the operator who selects the upstream display as the display mode of the image representing the blood vessel. An operation for displaying a partial region corresponding to a portion located upstream from the reference point on the image display device in a manner that is easier for the operator to recognize than other regions and selecting the downstream display as the display manner In accordance with the input by the person, the portion of the entire area representing the blood vessel in the image representing the blood vessel in the downstream side of the reference point of the branch where the set reference point of the graph structure is located and the direction of the blood flow by the determining means Provided with display control means for displaying on the image display device a partial area corresponding to the branch part determined to be away from the branch node in a manner that is easier for the operator to recognize than the other areas It may be.

  Here, as a specific example in which a specific partial region of the entire region representing the blood vessel is displayed on the image display device in a manner that is more easily recognized by the operator than other regions, for example, the specific region of the entire region representing the blood vessel is identified. For example, displaying only the partial area of the image on the image display device and excluding other areas from the display target, displaying the entire area representing the blood vessel in which the specific partial area is colored with a conspicuous color, etc. is there.

  Further, the image processing apparatus of the present invention includes a first partial region corresponding to a portion located upstream of the set reference point of the graph structure in the entire region representing the blood vessel in the image representing the blood vessel, A downstream portion of the branch where the set reference point is located in the graph structure and a second portion corresponding to the branch portion in which the direction of blood flow is determined to be away from the branch node by the determination means There may be provided display control means for displaying the area on the image display device in different colors so as to be distinguishable from each other.

  The display control means includes a third portion corresponding to a branch portion of the entire region representing the blood vessel in the image representing the blood vessel, the blood flow direction of which is determined by the determination means to be a direction toward the branch node. The partial area may be displayed on the image display device in a color different from the first partial area and the second partial area.

  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 apparatus, method, and program of the present invention, a reference point is set on a graph structure representing a blood vessel structure including a loop-shaped portion that has been acquired in advance in accordance with an input by an operator using the input device. And a downstream determination means for determining a downstream side of the blood flow from the reference point in a branch where the set reference point of the graph structure is located, and a downstream side of the determined blood flow from the reference point of the graph structure Arbitrary branch nodes that are located on the route that follows the path have directions based on the positions of a plurality of nodes that are located on the branch source branch, and have a direction toward the branch node. And a direction based on the positions of a plurality of nodes located on any one branch branched from the branch source branch, and on the side away from the branch node A second defined to have an orientation Using the angle θ (0 <θ <180 °) formed by the vector, the likelihood that the direction of the blood flow in one branch is away from the branch node is set to be larger as the angle θ is smaller. And using the calculated likelihood to determine whether the direction of blood flow in the one branch is away from the branch node or toward the branch node. In the blood vessel including the portion, the direction of blood flow in the partial blood vessel branched from the branch source branch can be specified.

  In the image processing apparatus, method, and program of the present invention, the determination processing is based on the likelihood based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to the plurality of nodes located on the branch source branch. The first representative value defined in the above is larger than the second representative value defined based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to the plurality of nodes located on the one branch. When the calculation is performed so that the value is as large as possible, in the blood vessel including the loop-shaped portion, the direction of the blood flow in the partial blood vessel branched from the branch source can be more accurately specified. .

The figure which shows schematic structure of the image processing apparatus which concerns on embodiment of this invention The figure for demonstrating the determination process by a determination means The figure for demonstrating the acquisition method of the 1st and 2nd vector by a determination means The figure for demonstrating the acquisition method of the 1st and 2nd representative value by a determination means The figure which shows an example of the determination result by a determination means The figure which shows an example of the image by the downstream display mode displayed by the display control means The figure for demonstrating an example of the blood vessel containing a loop-shaped part.

  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 shown in FIG. 1 displays an image representing a blood vessel including a loop-shaped portion such as a blood vessel of a large intestine, for example, for a doctor or the like to plan an operation, and observes the travel of the blood vessel and the direction of blood flow In this case, a display suitable for the observation is provided, and a graph extraction unit 10, a reference point setting unit 20, a downstream side determination unit 30, a determination unit 40, and a display control unit 50 are provided. Hereinafter, in the present embodiment, a case will be described in which the blood vessel including the loop-shaped portion in the present invention is a blood vessel of the large intestine.

  The graph extraction unit 10 extracts a graph structure representing a blood vessel from the three-dimensional image V in which the blood vessels of the large intestine are represented. Specifically, first, a region having image features and / or structural features of the blood vessels of the large intestine is extracted from the three-dimensional image V, and the thin line obtained by thinning the region is used as a branch point and a predetermined distance. A temporary graph structure is generated by dividing by. Here, the three-dimensional image V is composed of a large number of two-dimensional images acquired by, for example, a CT apparatus, an MRI apparatus, an ultrasonic diagnostic apparatus or the like stored in the data storage means.

  Next, a structure that matches or most closely resembles the shape model of the large intestine blood vessel stored in the shape model storage means DB is formed from 0 among a plurality of candidate points (nodes) constituting the generated temporary graph structure. Graph by selecting a plurality of candidate points as corresponding points corresponding to each of a plurality of teacher labels (nodes) constituting the shape model, and connecting the selected plurality of corresponding points so as to substantially match the shape model Generate and extract structures. Here, the shape model is a graph structure representing a general shape of a blood vessel of the large intestine, and can be acquired 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 (2). Here, R represents a set of correspondences between candidate points and teacher labels in the permissible solution x, θa represents a cost for an arbitrary correspondence a belonging to the set R, and θab represents two correspondences a belonging to the set R. Represents the cost for the combination of and b.

  The reference point setting means 20 is operated by the input device 2 such as a mouse or a keyboard for designating an arbitrary position on the three-dimensional image V displayed on the image display device 3 such as a display. A position on the three-dimensional image V corresponding to the designated position is specified in response to an input from the user, and if the specified position on the three-dimensional image V is on the extracted graph structure, If the specified position is not on the graph structure, the reference point S is set to the position on the graph structure closest to the specified position.

  The downstream side determination unit 30 determines the blood vessel among the two sides of the reference point based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to the plurality of nodes located before and after the set reference point S of the graph structure. The one whose average value of the diameter is smaller than the other is determined as the downstream side of the blood flow.

The determination unit 40 is configured to select each of the branches branched from the branch source branch at an arbitrary branch node B located on the path that follows the graph structure from the reference point S to the downstream side of the determined blood flow. The likelihood L which is the direction in which the direction of blood flow in each branch is away from the branch node B is calculated by the following formula (1), and the calculated likelihood When L is larger than a predetermined threshold, it is determined that the direction of blood flow in the branch is a direction away from the branch node B. When the likelihood L does not exceed the threshold, the direction of blood flow in the branch is branched. It is determined that the direction is toward node B.

  In the above formula (1), V1 represents a first vector having a direction based on the positions of a plurality of nodes located on the branch source branch and having a direction toward the branch node B, V2 represents a second vector having a direction based on the positions of a plurality of nodes located on a branch branched from the branch source branch and having a direction away from the branch node B, and F ( V1, V2) represents a function that increases as the angle θ formed by the first vector V1 and the second vector V2 decreases.

  R1 represents a representative value (for example, an average value) defined based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to a plurality of nodes located on the branch source branch, and R2 represents a branch. A representative value (for example, an average value) defined on the basis of the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to a plurality of nodes located on a branch branched from the original branch, G (R1, R2) ) Represents a function that increases as the difference obtained by subtracting the second representative value R2 from the first representative value R1 increases. W1 and W2 are coefficients for weighting the above functions.

  The threshold is a likelihood calculated when the angle θ formed by the first vector V1 and the second vector V2 is 90 ° or less and the first representative value is larger than the second representative value. Calculated when the angle θ formed by the first vector V1 and the second vector V2 is larger than 90 °, and the first representative value is smaller than the second representative value. A value larger than the value of likelihood L is set.

  For example, when determining the direction of blood flow in the branch E2 branched from the branch source branch E1 shown in FIG. 2, as shown in FIG. 3, five nodes P11 to P11 located on the branch source branch E1 are shown. The direction based on the position of P15 and the first vector V1 having the direction toward the branch node B is obtained, and the direction based on the positions of the five nodes P21 to P25 located on the branch E2 And the second vector V2 having a direction away from the branch node B is obtained, the first vector V1 and the second vector V2 are passed to the function F, and the above equation (1) is obtained. The value of the first term of is obtained. At this time, the value of the first term becomes larger as the angle β formed by the first vector V1 and the second vector V2 is smaller.

  Further, as shown in FIG. 4, the average values of the blood vessel diameters R11 to R15 at the plurality of positions on the blood vessel corresponding to the five nodes P11 to P15 located on the branch source branch E1 are the first representative. As the value R1, the average value of the blood vessel diameters R21 to R25 at each of a plurality of positions on the blood vessel corresponding to the five nodes P21 to P25 located on the branch E2 is acquired as the second representative value R2, The first representative value R1 and the second representative value R2 are passed to the function G to obtain the value of the second term of the above formula (1). At this time, the diameter of the blood vessel is obtained by obtaining the blood vessel region RV having image characteristics and / or structural characteristics of the blood vessels of the large intestine from the three-dimensional image V and obtaining the diameter of the cross section perpendicular to the traveling direction. (See JP 2010-220742 A). Then, the likelihood L is calculated by adding the values of the respective first and second terms obtained, the calculated likelihood L is compared with the threshold value, and the direction of blood flow in the branch E2 Determine.

  Further, when determining the direction of blood flow in the branch E3 branched from the branch source branch E1 shown in FIG. 2, as shown in FIG. 3, five nodes located on the branch source branch E1 are used. A first vector V1 having a direction based on the positions of P11 to P15 and having a direction toward the branch node B is obtained, and at the positions of five nodes P31 to P35 located on the branch E3. A second vector V3 having a direction based on and a direction away from the branch node B is obtained, and the first vector V1 and the second vector V3 are set to the variable V1 in the equation 1 is assigned to the function F (so that the second vector V3 is assigned to the variable V2) and the value of the first term of the above equation (1) is obtained. At this time, the value of the first term becomes larger as the angle α formed by the first vector V1 and the second vector V3 is smaller.

  Further, as shown in FIG. 4, the average values of the blood vessel diameters R11 to R15 at the plurality of positions on the blood vessel corresponding to the five nodes P11 to P15 located on the branch source branch E1 are the first representative. As the value R1, the average value of the blood vessel diameters R31 to R35 at each of a plurality of positions on the blood vessel corresponding to the five nodes P31 to P35 located on the branch E3 is acquired as the second representative value R3, A function of the first representative value R1 and the second representative value R3 (so that the first representative value R1 is substituted for the variable R1 and the second representative value R3 is substituted for the variable R2). The value of the second term of the above formula (1) is obtained by passing to G. Then, the likelihood L is calculated by adding the values of the respective first and second terms obtained, the calculated likelihood L is compared with the threshold value, and the direction of blood flow in the branch E2 Determine.

  In addition, as a result of the determination using the likelihood L for each of a plurality of branches branched from the branch source branch, the determination unit 40 determines that all the directions of blood flow in the plurality of branches are branch nodes B. When it is determined that the direction is the direction of heading, it is determined that the direction of blood flow in one branch having the highest likelihood L calculated for the branch among a plurality of branches is a direction away from the branch node B. It has a function to do.

  For example, as a result of the determination using the likelihood L for each of the two branches E2 and E3 branched from the branch source branch E1 shown in FIG. When it is determined that the direction is toward the branch node B, the direction of blood flow in the branch having the larger likelihood L calculated for the branch among the branches E2 and E3 is a direction away from the branch node B. Judge that there is.

  The display control means 50 controls the display of an image representing a blood vessel such as the three-dimensional image V in the image display device 3 such as a display. For example, the display control means 50 determines the result of the determination by the determination means 40 as shown in FIG. Based on the first partial region RU corresponding to the portion located upstream of the reference point S of the graph structure, and the branch where the reference point S of the graph structure is located A second partial region RD corresponding to a portion downstream from the reference point S and a branch portion whose direction of blood flow is determined to be away from the branch node B by the determination means 40; The direction of the blood flow is divided into three regions of the third partial region RB corresponding to the branch portion determined to be the direction toward the branch node B (see FIG. 6), and the first partial region RU. And the second partial region RD Display an image on the image display device 3 in which different regions or three regions of the first partial region RU, the second partial region RD, and the third partial region RB are displayed in different colors. .

  In addition, the display control unit 50 can recognize the first partial region RU more easily than the second partial region RD and the third partial region RB in the entire region representing the blood vessels of the large intestine in the image. And an upstream display mode for displaying the second partial area RD in a manner that is easier for the operator to recognize than the first partial area RU and the third partial area RB, A function of displaying an image in the selected display mode on the image display device 3 in response to an input of an operator who selects one of the display modes by the input device 2 such as a mouse or a keyboard. For example, in response to an input from the operator who selects the downstream display mode, the first partial region RU and the third partial region RB as shown in FIG. 6 are excluded from the display target, and the second partial region RD. An image displaying only the image is displayed on the image display device 3.

  Further, the display control means 50 applies the fourth partial area, which is a set of the second partial area RD and the partial area located within a predetermined range from the branch node B of the third partial area RB, to the other partial areas. It is also provided with a function of displaying in a manner that is easier for the operator to recognize than the area. This fourth partial region corresponds to a region to be excised in the excision operation for excising the blood vessel portion downstream from the reference point S, and the fourth partial region is connected to a region representing another blood vessel. Since the position corresponds to the position where the blood flow should be stopped with a clip when performing resection surgery, providing such an indication supports the planning operation of resection surgery by an operator such as a doctor. The work efficiency can be improved.

  With the above configuration, in the image processing apparatus 1, first, the 3D image V stored in the data storage unit is read in response to a request from the operator, and the graph extraction unit 10 represents the blood vessel from the 3D image V. The graph structure is extracted, and the display control means 50 displays the three-dimensional image V on the image display device 3. Thereafter, the reference point setting means 20 receives an input from an operator who designates an arbitrary position on the three-dimensional image V, and sets the reference point S at a position on the graph structure corresponding to the position designated by the input.

  After that, when the direction of blood flow in the branch where the reference point S is located is specified by a method described later, the determination unit 40 follows the graph structure downstream from the reference point S and branches on the path. The direction of blood flow in each branch branched from the branch source branch is sequentially determined. The determination means 40 newly identifies a branch whose direction of blood flow is away from the branch node by the determination, extends the path so as to pass through the branch, and branches on the extended path. The process of further determining at the node is repeated. As a result, the direction of blood flow in each of the branches branched from the branch source branch is determined at all branch nodes located on the path from the reference point S to the peripheral part of the graph structure.

  Thereafter, the display control unit 50 determines the regions representing the blood vessels in the image based on the determination result by the determination unit 40 as the first partial region RU, the second partial region RD, and the third partial region RB. An image is displayed on the image display device 3 according to a display mode set in advance as an initial display, for example, an image in which the regions are displayed in different colors so that they can be distinguished from each other. The display control means 50 further detects whether or not an input by an operator who selects a screen display in a display mode different from the image being displayed is made, and when the input is made, the image according to the selected display mode. Switch the display to.

  Here, the direction of blood flow in the branch where the reference point S is located is the direction in which the diameter of the blood vessel becomes narrower by examining the diameter of the blood vessel at nodes 5 to 10 before and after the reference point S, respectively. The direction of blood flow in the branch may be specified, or the operator may input information on which of the branches is upstream, and may be specified based on the information.

  In the above-described embodiment, the case is described in which the determination unit 40 performs the determination based on information on both the branch angle and the blood vessel diameter. However, based on only the information on the branch angle. The determination may be performed. For example, when determining the direction of blood flow in the branch E2 branched from the branch source branch E1 shown in FIG. 2, if the angle β formed by the first vector V1 and the second vector V2 is greater than 90 degrees, When it is determined that the direction of blood flow in the branch E2 is a direction toward the branch node B and the angle θ does not exceed 90 degrees, the direction of blood flow in the branch E2 is a direction away from the branch node. Can be determined.

  In the above embodiment, the case where the graph extracting unit 10 directly extracts a graph structure representing a blood vessel by performing image analysis on the three-dimensional image V has been described. However, the graph structure has already been extracted and recorded. If the data is recorded on the medium, the data having the graph structure may be read from the recording medium and provided to the subsequent processing without being extracted separately.

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 10 Graph extraction means 20 Reference point setting means 30 Downstream side determination means 40 Determination means 50 Display control means S Reference point B Branch node E1 Branch source branch E2, E3 Branch branch V1 First vector V2, V3 Second vector RU First partial region RD Second partial region RM Third partial region

Claims (12)

A reference point setting means for setting a reference point on a graph structure representing a blood vessel structure including a loop-shaped portion acquired in advance according to an input from an operator by an input device;
In the branch where the set reference point of the graph structure is located, downstream determination means for determining the downstream side of the blood flow from the reference point;
A direction based on the positions of a plurality of nodes located on a branch source branch in an arbitrary branch node located on a path following the determined blood flow downstream from the reference point of the graph structure And a first vector defined to have a direction toward the branch node and a plurality of branches located on any one branch branched from the branch source branch Using an angle θ formed by a second vector having a direction based on the position of the node and having a direction away from the branch node, the blood flow in the one branch is The likelihood that the direction is a direction away from the branch node is calculated so as to be larger as the angle θ is smaller, and the direction of blood flow in the one branch is calculated using the calculated likelihood. A direction away from the branch node or the branch node An image processing apparatus comprising: a determination unit that determines whether the direction is toward the camera.   The determination means determines the likelihood based on a diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to a plurality of nodes located on the branch source branch. Is larger as it is larger than a second representative value defined based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to a plurality of nodes located on the one branch. The image processing apparatus according to claim 1, wherein the image processing apparatus calculates a value.   As a result of the determination using the likelihood for each of a plurality of branches branched from the branch source branch, the directions of blood flow in the plurality of branches are all directed to the branch node. If it is determined that the direction is a direction, it is determined that the direction of blood flow in the one branch having the highest likelihood calculated for the branch among the plurality of branches is a direction away from the branch node. The image processing apparatus according to claim 1, wherein the image processing apparatus is one.   When the determination unit determines that the direction of blood flow in the one branch is a direction toward the branch node when the angle θ is greater than a predetermined threshold, and the angle θ does not exceed the threshold The image processing apparatus according to claim 1, wherein the direction of blood flow in the one branch is determined to be a direction away from the branch node.   Based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to a plurality of nodes positioned before and after the reference point, the downstream side determination unit determines the diameter of the blood vessel among both sides of the reference point. The image processing apparatus according to any one of claims 1 to 4, wherein an average value smaller than the other is determined as a downstream side of the blood flow. Upstream from the set reference point of the graph structure in the entire region representing the blood vessel in the image representing the blood vessel in response to an input by an operator who selects upstream display as a display mode of the image representing the blood vessel. The partial region corresponding to the portion located on the side is displayed on the image display device in a manner that is more easily recognized by the operator than other regions, and input by the operator who selects the downstream display as the display manner. Accordingly, a portion of the entire region representing the blood vessel in the image representing the blood vessel that is downstream of the reference point of the branch where the set reference point of the graph structure is located and the blood flow of the blood flow by the determination unit. A partial area corresponding to a branch portion determined to be away from the branch node is displayed on the image display device in a manner that is easier for the operator to recognize than the other areas. The image processing apparatus according to any one of claims 1 to 5, characterized in that a display control unit. Of the entire region representing the blood vessel in the image representing the blood vessel, a first partial region corresponding to a portion located upstream from the set reference point of the graph structure, and the set of the graph structure A downstream portion of the branch where the reference point is located and a second partial region corresponding to the branch portion in which the direction of blood flow is determined to be away from the branch node by the determination means. 6. The image processing apparatus according to claim 1, further comprising display control means for displaying on the image display apparatus in different colors so as to be distinguishable.   The display control means corresponds to a branch portion of the entire region representing the blood vessel in the image representing the blood vessel, the direction of the blood flow being determined by the determination means to be a direction toward the branch node. The image according to claim 7, wherein the third partial area to be displayed is displayed on the image display device in a color different from the first partial area and the second partial area. Processing equipment. A process of setting a reference point on a graph structure representing a blood vessel structure including a loop-shaped portion acquired in advance according to an input of an operator by an input device;
In the branch where the set reference point of the graph structure is located, a process of determining the downstream side of the blood flow from the reference point;
A direction based on the positions of a plurality of nodes located on a branch source branch in an arbitrary branch node located on a path following the determined blood flow downstream from the reference point of the graph structure And a first vector defined to have a direction toward the branch node and a plurality of branches located on any one branch branched from the branch source branch Using an angle θ formed by a second vector having a direction based on the position of the node and having a direction away from the branch node, the blood flow in the one branch is The likelihood that the direction is a direction away from the branch node is calculated so as to be larger as the angle θ is smaller, and the direction of blood flow in the one branch is calculated using the calculated likelihood. A direction away from the branch node or the branch node An image processing method comprising: executing a process for determining whether the direction is directed to at least one computer.   In the determination process, the likelihood is determined based on a diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to a plurality of nodes located on the branch source branch. Is larger as it is larger than a second representative value defined based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to a plurality of nodes located on the one branch. The image processing method according to claim 9, wherein the calculation is performed. Computer
A reference point setting means for setting a reference point on a graph structure representing a blood vessel structure including a loop-shaped portion acquired in advance according to an input from an operator by an input device;
In the branch where the set reference point of the graph structure is located, downstream determination means for determining the downstream side of the blood flow from the reference point;
A direction based on the positions of a plurality of nodes located on a branch source branch in an arbitrary branch node located on a path following the determined blood flow downstream from the reference point of the graph structure And a first vector defined to have a direction toward the branch node and a plurality of branches located on any one branch branched from the branch source branch The angle θ formed by the second vector having a direction based on the position of the node and having a direction on the side away from the branch node is used to determine the blood flow in the one branch. The likelihood that the direction is a direction away from the branch node is calculated so as to be larger as the angle θ is smaller, and the direction of blood flow in the one branch is calculated using the calculated likelihood. A direction away from the branch node or the branch node An image processing program for functioning as a determination unit for determining whether the direction is toward the camera.   The determination means determines the likelihood based on a diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to a plurality of nodes located on the branch source branch. Is larger as it is larger than a second representative value defined based on the diameter of the blood vessel at each of a plurality of positions on the blood vessel corresponding to a plurality of nodes located on the one branch. 12. The image processing program according to claim 11, wherein the image processing program is calculated.