JP7483591B2 - Medical image processing device, medical image processing method and program - Google Patents

Description

The embodiments disclosed in this specification and the drawings relate to a medical image processing device, a medical image processing method, and a program.

In a patient's body, there are various organs through which fluids or solids flow, such as biological organs with tubular structures, such as blood vessels and the esophagus, or valves. Information such as measured values in these organs is useful for making diagnoses. However, these organs are not necessarily simply cylindrical, and it is not easy to obtain information such as measured values.

JP 2004-283373 A JP 2020-76967 A

One of the problems that the embodiments disclosed in this specification and the drawings attempt to solve is to provide information about a biological organ or valve having a tubular structure. However, the problems that the embodiments disclosed in this specification and the drawings attempt to solve are not limited to the above problem. Problems that correspond to the effects of each configuration shown in the embodiments described below can also be positioned as other problems.

The medical image processing device of the embodiment includes an acquisition unit, an axis setting unit, a condition setting unit, and an identification unit. The acquisition unit acquires a medical image including a blood vessel or a valve. The axis setting unit sets a plurality of axes based on the blood vessel or the valve. The condition setting unit sets a condition related to a closed curve corresponding to the blood vessel or the valve. The identification unit identifies a closed curve based on the plurality of axes or a region surrounded by the closed curve based on the condition.

FIG. 1 is a block diagram showing an example of the configuration of a medical image processing system according to the first embodiment. FIG. 2 is a flowchart showing an example of processing by the processing circuitry of the medical image processing apparatus according to the first embodiment. FIG. 3 is a diagram illustrating an example of a region of interest according to the first embodiment. FIG. 4A is a flowchart illustrating an example of a process related to axis setting in the axis setting function according to the first embodiment. FIG. 4B is a diagram for explaining the axis setting method according to the first embodiment. FIG. 4C is a diagram for explaining a method of setting an axis according to the first embodiment. FIG. 4D is a diagram for explaining the axis setting method according to the first embodiment. FIG. 5A is a diagram illustrating an example of a method for setting endpoints according to the first embodiment. FIG. 5B is a diagram illustrating an example of a method for setting endpoints according to the first embodiment. FIG. 6A is a diagram for explaining the setting of a closed curve condition according to the first embodiment. FIG. 6B is a diagram for explaining the setting of the closed curve condition according to the first embodiment. FIG. 7A is a diagram for explaining setting of evaluation points according to the first embodiment. FIG. 7B is a diagram for explaining setting of the evaluation points according to the first embodiment. FIG. 8A is a diagram showing a display example according to the first embodiment. FIG. 8B is a diagram showing a display example according to the first embodiment. FIG. 8C is a diagram showing a display example according to the first embodiment. FIG. 8D is a diagram showing a display example according to the first embodiment. FIG. 9 is a diagram showing a display example according to the first embodiment. FIG. 10 is a diagram illustrating an example of a stenotic site in a blood vessel according to the first embodiment. FIG. 11 is a diagram showing a display example according to the first embodiment. FIG. 12 is a diagram showing an example of a method for setting an axis according to the first embodiment. FIG. 13A is a diagram showing an example of a method for setting a closed curve condition according to the first embodiment. FIG. 13B is a diagram showing an example of a method for setting a closed curve condition according to the first embodiment. FIG. 14 is a diagram for explaining an axis setting method according to the fourth embodiment.

Below, the embodiments of the medical image processing device, medical image processing method, and program will be described in detail with reference to the attached drawings.

First Embodiment
In this embodiment, a medical image processing system 1 including a medical image processing device 20 will be described as an example. For example, as shown in Fig. 1, the medical image processing system 1 has a medical image diagnostic device 10, a medical image processing device 20, and an image storage device 30. Fig. 1 is a block diagram showing an example of the configuration of the medical image processing system 1 according to the first embodiment. The medical image diagnostic device 10, the medical image processing device 20, and the image storage device 30 are connected to each other via a network NW.

In addition, as long as they can be connected via the network NW, the locations where each device included in the medical image processing system 1 is installed are arbitrary. For example, the medical image diagnostic device 10, the medical image processing device 20, and the image storage device 30 may be installed in different facilities. In other words, the network NW may be configured as a closed local network within the facility, or may be a network via the Internet.

The medical image diagnostic device 10 is a device that captures images of a patient and collects medical images including a biological organ or valve having a tubular structure. In this embodiment, as an example, a case in which a medical image including the mitral valve is collected will be described. That is, in this embodiment, the mitral valve will be described as the target organ.

The medical image diagnostic device 10 is, for example, a medical modality such as an X-ray diagnostic device, an X-ray CT (Computed Tomography) device, an MRI (Magnetic Resonance Imaging) device, an ultrasound diagnostic device, a SPECT (Single Photon Emission Computed Tomography) device, or a PET (Positron Emission computed Tomography) device. Although a single medical image diagnostic device 10 is shown in FIG. 1, the medical image processing system 1 may include multiple medical image diagnostic devices 10. The medical image processing system 1 may also include multiple types of medical image diagnostic devices 10. For example, the medical image processing system 1 may include an X-ray CT device and an MRI device as the medical image diagnostic devices 10.

The image storage device 30 is an image database that stores medical images collected by the medical image diagnostic device 10. For example, the image storage device 30 includes an arbitrary storage device inside or outside the device, and manages medical images acquired from the medical image diagnostic device 10 via the network NW in the form of a database. For example, the image storage device 30 is a PACS (Picture Archiving and Communication System) server. The image storage device 30 may also be realized by a group of servers (cloud) connected to the medical image processing system 1 via the network NW.

The medical image processing device 20 is a device that performs various processes described below to enable the provision of information about biological organs or valves having a tubular structure. For example, the medical image processing device 20 acquires a medical image including a mitral valve, sets multiple axes based on the mitral valve, sets conditions related to a closed curve corresponding to the mitral valve, and identifies a closed curve based on the multiple axes set or an area surrounded by the closed curve based on the set conditions. For example, as shown in FIG. 1, the medical image processing device 20 includes a memory 21, a display 22, an input interface 23, and a processing circuit 24.

The memory 21 is realized, for example, by a semiconductor memory element such as a RAM (Random Access Memory), a flash memory, a hard disk, an optical disk, etc. For example, the memory 21 stores medical images collected by the medical image diagnostic device 10. The memory 21 also stores programs that enable the circuits included in the medical image processing device 20 to realize their functions.

The display 22 displays various information. For example, the display 22 displays a GUI (Graphical User Interface) for receiving various instructions and settings from the user via the input interface 23. The display 22 also displays information such as a closed curve or area identified by the processing circuitry 24, or measurement values acquired based on the closed curve or area. For example, the display 22 is a liquid crystal display or a CRT (Cathode Ray Tube) display. The display 22 may be a desktop type, or may be configured as a tablet terminal or the like capable of wireless communication with the medical image processing device 20 main body.

In FIG. 1, the medical image processing device 20 is described as having a display 22, but the medical image processing device 20 may have a projector instead of or in addition to the display 22. The projector can project onto a screen, wall, floor, the body surface of a patient, etc. under the control of the processing circuitry 24. As an example, the projector can also project onto any plane, object, space, etc. by projection mapping.

The input interface 23 accepts various input operations from the user, converts the accepted input operations into electrical signals, and outputs them to the processing circuit 24. For example, the input interface 23 is realized by a mouse, keyboard, trackball, switch, button, joystick, a touchpad that performs input operations by touching the operation surface, a touch screen in which a display screen and a touchpad are integrated, a non-contact input circuit using an optical sensor, a voice input circuit, etc. The input interface 23 may be configured as a tablet terminal or the like that can wirelessly communicate with the medical image processing device 20 main body. The input interface 23 may also be a circuit that accepts input operations from the user by motion capture. For example, the input interface 23 can accept the user's body movements, gaze, etc. as input operations by processing signals acquired via a tracker and images collected about the user. The input interface 23 is not limited to only those that have physical operating parts such as a mouse and a keyboard. For example, an example of the input interface 23 includes an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the medical image processing device 20 and outputs this electrical signal to the processing circuit 24.

The processing circuitry 24 controls the operation of the entire medical image processing device 20 by executing a control function 24a, an acquisition function 24b, an axis setting function 24c, a condition setting function 24d, a specification function 24e, and an output function 24f. The acquisition function 24b is an example of an acquisition section. The axis setting function 24c is an example of an axis setting section. The condition setting function 24d is an example of a condition setting section. The specification function 24e is an example of a specification section. The output function 24f is an example of an output section.

For example, the processing circuit 24 reads out from the memory 21 a program corresponding to the control function 24a and executes it to control various functions such as the acquisition function 24b, the axis setting function 24c, the condition setting function 24d, the identification function 24e, and the output function 24f based on various input operations received from the user via the input interface 23.

The processing circuitry 24 also acquires medical images including the mitral valve by reading out from the memory 21 a program corresponding to the acquisition function 24b and executing it. For example, the acquisition function 24b receives medical images captured by the medical image diagnostic device 10 via the network NW and stores them in the memory 21. Here, the acquisition function 24b may acquire medical images directly from the medical image diagnostic device 10, or may acquire medical images via the image storage device 30.

The processing circuit 24 also sets multiple axes by reading from the memory 21 a program corresponding to the axis setting function 24c and executing it. The processing circuit 24 also sets conditions related to the closed curve by reading from the memory 21 a program corresponding to the condition setting function 24d and executing it. The processing circuit 24 also specifies a closed curve based on the multiple axes that have been set or an area surrounded by the closed curve based on the set conditions by reading from the memory 21 a program corresponding to the specification function 24e and executing it. The processing circuit 24 also reads from the memory 21 a program corresponding to the output function 24f and executes it to perform output based on the specified closed curve or area. Details of the processing by the axis setting function 24c, the condition setting function 24d, the specification function 24e, and the output function 24f will be described later.

In the medical image processing device 20 shown in FIG. 1, each processing function is stored in the memory 21 in the form of a program executable by a computer. The processing circuitry 24 is a processor that realizes the function corresponding to each program by reading the program from the memory 21 and executing it. In other words, the processing circuitry 24 in a state in which a program has been read has the function corresponding to the read program.

In FIG. 1, the control function 24a, acquisition function 24b, axis setting function 24c, condition setting function 24d, specific function 24e, and output function 24f are described as being realized by a single processing circuit 24, but the processing circuit 24 may be configured by combining multiple independent processors, and each processor may execute a program to realize the functions. Furthermore, each processing function possessed by the processing circuit 24 may be realized by being appropriately distributed or integrated into a single or multiple processing circuits.

The processing circuitry 24 may also realize functions by using a processor of an external device connected via the network NW. For example, the processing circuitry 24 reads out and executes a program corresponding to each function from the memory 21, and realizes each function shown in FIG. 1 by using a group of servers (cloud) connected to the medical image processing device 20 via the network NW as a computational resource.

The above describes an example of the configuration of the medical image processing system 1 including the medical image processing device 20. With this configuration, the processing circuitry 24 in the medical image processing device 20 is capable of providing information about the mitral valve.

The processing performed by the processing circuitry 24 will be described below with reference to the flowchart in FIG. 2. FIG. 2 is a flowchart showing an example of processing performed by the processing circuitry 24 of the medical image processing apparatus 20 according to the first embodiment.

First, the acquisition function 24b acquires a medical image including the mitral valve (step S1). The acquisition function 24b may acquire the medical image directly from the medical image diagnostic device 10, or may acquire the medical image via another device such as the image storage device 30.

The type of medical image is not particularly limited. For example, the acquisition function 24b acquires X-ray CT images, ultrasound images, MRI images, X-ray images, PET images, SPECT images, etc. as medical images including the mitral valve. In addition, the acquisition function 24b can acquire any type of image in which morphological information of the three-dimensional anatomical structure of the mitral valve is stored as a medical image including the mitral valve. The acquisition function 24b may also acquire a four-dimensional image obtained by capturing multiple three-dimensional images of the mitral valve in the time direction as a medical image including the mitral valve.

As one example, the acquisition function 24b acquires medical images when triggered by an instruction received from a user via the input interface 23. As another example, the acquisition function 24b acquires medical images when triggered by a medical image including the mitral valve being captured by the medical image diagnostic device 10, or when a medical image including the mitral valve is stored in the image storage device 30. That is, the acquisition function 24b may automatically acquire newly collected medical images.

The acquisition function 24b may also acquire a newly collected medical image when the medical image satisfies a predetermined condition. For example, the acquisition function 24b may set an imaging protocol as a predetermined condition, and acquire the medical image when the image is captured using an imaging protocol that targets the heart. For example, the acquisition function 24b may set a reconstruction method as a predetermined condition, and acquire the medical image when enlarged reconstruction is performed. The acquisition function 24b may also combine multiple items, such as an imaging protocol and a reconstruction method, to set a predetermined condition, and acquire a medical image when the predetermined condition is satisfied.

Next, the axis setting function 24c acquires a region of interest in the medical image acquired by the acquisition function 24b (step S2). Here, the region of interest is a region indicated by a biological organ of interest in the medical image. For example, the axis setting function 24c acquires coordinate information of each pixel indicating the mitral valve in the medical image as the region of interest. That is, the axis setting function 24c acquires a region of interest from the medical image based on the anatomical characteristics of the mitral valve.

As an example, the output function 24f displays a medical image on the display 22. Then, the axis setting function 24c acquires the region of interest by receiving an operation from the user via the input interface 23 to specify the position of the region of interest.

As another example, the axis setting function 24c acquires a region of interest based on an anatomical structure depicted in a medical image using a known region extraction technique. Examples of known region extraction techniques include discriminant analysis based on pixel values such as CT values (also known as Otsu's binarization method), region growing method, snake method, graph cut method, and mean shift method.

Additionally, the axis setting function 24c can acquire the area of interest using any method. For example, the axis setting function 24c can acquire the area of interest using machine learning techniques such as deep learning. As one example, the axis setting function 24c may acquire the area of interest using a shape model of the area of interest that is constructed based on learning data prepared in advance.

The axis setting function 24c may first acquire an area (hereinafter, referred to as a related area) that is larger than the region of interest but smaller than the entire image, and then acquire the region of interest from the related area. For example, when the mitral valve is the target organ, the axis setting function 24c acquires the cardiac area or the combined area of the left atrium and left ventricle from the entire image as the related area. For example, the axis setting function 24c can acquire the related area by accepting an operation from the user via the input interface 23. Then, the axis setting function 24c applies a process such as a graph cut method to the related area to acquire the region of interest. This can reduce the calculation cost compared to when a process such as a graph cut method is performed on the entire image. Furthermore, by limiting the processing target to the related area, it becomes possible to acquire the region of interest using a method with a higher calculation load, and therefore the region of interest can be acquired with higher accuracy.

For example, the acquisition function 24b acquires the medical image I1 shown in FIG. 3. Furthermore, the axis setting function 24c acquires the mitral valve region R1 from the medical image I1 based on the mitral valve. That is, the axis setting function 24c acquires the mitral valve region R1 from the medical image I1 based on the anatomical characteristics of the mitral valve. The mitral valve region R1 is an example of a region of interest. Furthermore, FIG. 3 is a diagram showing an example of a region of interest according to the first embodiment.

Next, the axis setting function 24c sets multiple axes based on the mitral valve (step S3). That is, the axis setting function 24c sets multiple axes based on the anatomical characteristics of the mitral valve. For example, the axis setting function 24c sets multiple axes based on the mitral valve region R1 in FIG. 3. That is, the axis setting function 24c sets multiple axes along the shape of the acquired mitral valve region R1.

Hereinafter, the setting of the axes by the axis setting function 24c will be described with reference to FIGS. 4A to 4D.
Fig. 4A is a flowchart showing an example of a process related to axis setting in the axis setting function 24c according to the first embodiment. Note that steps S31, S32, and S33 in Fig. 4A are steps included in step S3 in Fig. 2. Figs. 4B, 4C, and 4D are diagrams for explaining the axis setting method according to the first embodiment.

First, the axis setting function 24c identifies the tip shape D1 and the annulus shape D2 as shown in FIG. 4B (step S31). The tip shape D1 is, for example, a region (pixels) indicating the tip of the valve identified as a closed curve. The annulus shape D2 is a region (pixels) indicating the annulus identified as a closed curve. Note that, although the tip shape D1 and the annulus shape D2 are shown as two-dimensional closed curves in FIG. 4B, the tip shape D1 and the annulus shape D2 are usually three-dimensional closed curves that vary in the depth direction. The tip shape D1 is an example of a first closed curve. The annulus shape D2 is an example of a second closed curve.

The method for identifying the tip shape D1 and the annulus shape D2 is not particularly limited. For example, the axis setting function 24c can identify the tip shape D1 and the annulus shape D2 based on the number of pixels belonging to the region of interest adjacent to each pixel. In other words, the axis setting function 24c can identify a region adjacent to a large number of pixels belonging to a region other than the region of interest as the tip shape D1 and the annulus shape D2.

As another example, the axis setting function 24c may identify the pixel located at the bottom or top of each pixel row (voxel row or pixel row) in a predetermined direction as the tip shape D1 and the annulus shape D2. Here, the predetermined direction is the direction from the valve tip to the annulus or the direction from the annulus to the valve tip.

For example, the predetermined direction may be determined based on image-related information such as a DICOM (Digital Imaging and Communications in Medicine) header, or may be specified by the user using the input interface 23. When the user specifies the predetermined direction, the UI (User Interface) may be configured to allow the user to directly specify the direction, or to set the direction from the front to the back of the display screen as the predetermined direction. As an example, the output function 24f displays the mitral valve region R1, which is three-dimensional data, on the display 22 in a rotatable manner. Also, the user rotates the mitral valve region R1 so that the opening of the mitral valve is oriented in a visible direction, as shown in, for example, Figures 4B to 4D. Then, the axis setting function 24c sets the direction from the front to the back of the display 22 as the predetermined direction based on the orientation of the mitral valve region R1 after rotation.

Next, the axis setting function 24c sets end points for each closed curve, for example as shown in FIG. 4C (step S32). For example, the axis setting function 24c sets the same number of end points for each of the tip shape D1 and the annulus shape D2 based on a predetermined condition. That is, the axis setting function 24c sets ten end points, end points E101 to E110, for the tip shape D1, and ten end points, end points E201 to E210, for the annulus shape D2. The predetermined condition for setting the end points may be a fixed condition, or may be set each time.

The predetermined conditions for setting the end points can be set, for example, using the GUI shown in FIG. 5A. FIG. 5A is a diagram showing an example of a method for setting end points according to the first embodiment. First, the output function 24f displays the GUI shown in FIG. 5A on the display 22. Next, the axis setting function 24c accepts input of each item shown in FIG. 5A from the user via the input interface 23. That is, the axis setting function 24c can set multiple end points for each of the tip shape D1 or the annulus shape D2 by accepting input of a starting point for sequentially setting multiple end points and conditions for setting multiple end points using the GUI shown in FIG. 5A.

For example, the user inputs the directions (vectors) of the "tip" and "annulus" as starting points. If the "link tip and annulus" item is checked, the axis setting function 24c reflects the direction input for either the "tip" or the "annulus" to the other. For example, if the starting point direction of the "tip" and the "annulus" is "90 degrees", the axis setting function 24c sets the starting points E111 and E211 at a position "90 degrees" from the reference point, as shown in FIG. 5B. The starting point E111 is one of the end points set for the tip shape D1, and the starting point E211 is one of the end points set for the annulus shape D2. FIG. 5B is a diagram showing an example of a method for setting end points according to the first embodiment. In FIG. 5B, the upward direction of the screen is described as the reference direction (0 degrees), but the embodiment is not limited to this, and the reference direction may be determined based on the center of gravity or center of the tip shape D1 or the annulus shape D2, for example.

The method for setting the reference point is arbitrary. For example, the reference point may be the center of the display screen, or a coordinate point indicating the center or center of gravity of the tip shape D1 or the annulus shape D2. Furthermore, when the tip shape D1 and the annulus shape D2 are three-dimensional closed curves, the reference point may be the position where the coordinate point indicating the center or center of gravity of the tip shape D1 or the annulus shape D2 is projected perpendicularly onto the cross section displayed on the display 22. Furthermore, the reference point may be specified by the user.

After setting the starting point, the axis setting function 24c sets multiple end points according to the "Condition" setting in FIG. 5A. For example, if "Number: 10" is set as shown in FIG. 5A, the axis setting function 24c sets the remaining nine end points on the tip shape D1 at regular intervals from the starting point E111. Note that in this case, the longer the tip shape D1 is, the longer the interval between the end points will be. Similarly, the axis setting function 24c sets the remaining nine end points on the annulus shape D2 at regular intervals from the starting point E211.

Note that, although a case where multiple end points are set at regular intervals has been described, the embodiment is not limited to this. For example, the axis setting function 24c may set multiple end points based on the starting point position, the set number, and shape information of the closed curve. For example, the axis setting function 24c may calculate the curvature at each position of the closed curve as shape information of the closed curve, and set a set number of end points in order from the position with the largest or smallest curvature.

Although the case where the starting point is set based on a reference point and a direction (vector) has been described, the starting point may be set manually by the user. For example, the user can set the starting point at any position on the tip shape D1 or the annulus shape D2 via the input interface 23. For example, the output function 24f may display the starting point at any position on the tip shape D1 or the annulus shape D2, and the user may move the position of the starting point by operating the mouse pointer M shown in FIG. 5B. That is, the axis setting function 24c may accept a user operation on the display of the tip shape D1 or the annulus shape D2 as input of the starting point. Alternatively, the starting point may be set in advance. For example, the starting point may always be set to a position toward the top of the screen.

Although FIG. 5A describes a case where the number of end points is set, the "tip end point interval" or the "annulus end point interval" may be set. In other words, the axis setting function 24c may set the minimum distance or the number of pixels between adjacent end points in the tip shape D1 or the annulus shape D2.

When "tip end point interval" is set, axis setting function 24c sets multiple end points on tip shape D1 at the set interval from the starting point. For example, axis setting function 24c calculates the maximum number of end points that can be set at equal intervals and with a distance between each end point smaller than the set distance based on the length of tip shape D1, sets all end points on tip shape D1, and sets the same number of end points at equal intervals on annulus shape D2. Similarly, when "annulus end point interval" is set, multiple end points can be set on tip shape D1 and annulus shape D2.

When setting the end points, the axis setting function 24c associates the end points on the tip shape D1 with the end points on the annulus shape D2, and records the correspondence. For example, the axis setting function 24c can record the correspondence by assigning numbers in sequence clockwise or counterclockwise from the starting point, based on the starting point.

After setting the end points, the axis setting function 24c may accept modifications from the user. For example, as shown in FIG. 4C, after multiple end points are set for each of the tip shape D1 and the annulus shape D2, the output function 24f displays these multiple end points on the display 22. The output function 24f may display the multiple end points in association with an image or schematic diagram showing the mitral valve. The axis setting function 24c then accepts modifications related to the end points via the input interface 23. For example, the user can use a mouse or the like to move the end points, delete some end points, or add end points.

When accepting modifications to the end points, the axis setting function 24c may control the end points on the tip shape D1 and the end points on the annulus shape D2 to move in conjunction with each other. For example, when an end point on the tip shape D1 is moved, the axis setting function 24c may control the corresponding end point on the annulus shape D2 to move in conjunction with each other.

Furthermore, when accepting modifications to the end points, the axis setting function 24c may impose restrictions on the modifications. For example, when the user moves an end point, the axis setting function 24c may control so that it cannot be moved beyond a set condition. For example, when a minimum distance between end points is set, the axis setting function 24c controls so that it cannot be moved beyond that minimum distance. Furthermore, when the number of end points is set, the axis setting function 24c may control so that it cannot be deleted or added to an end point.

Next, the axis setting function 24c sets an axis to pass through the corresponding end points, for example as shown in FIG. 4D (step S33). For example, the axis setting function 24c sets the axis A101 based on the end point E101 set on the tip shape D1 and the end point E201 corresponding to the end point E101 among the multiple end points set on the annulus shape D2. For example, the axis setting function 24c sets the line segment that passes through the end points E101 and E201 and has the shortest distance along the shape of the mitral valve region R1 as the axis A101. In a similar manner, the axis setting function 24c sets the axes A102 to A110.

In FIG. 4C, the same number of end points are set for each of the tip shape D1 and the annulus shape D2, but the axis setting function 24c may set different numbers of end points for the tip shape D1 and the annulus shape D2. Also, it is not necessary to provide a correspondence between the end points.

That is, the axis setting function 24c may set many end points on one closed curve and not use all of the end points, or may set multiple axes passing through one end point. The axis setting function 24c may also set all combinations as axes. The axes that are set may intersect. However, as shown in FIG. 5A, etc., when an axis is set in step S3 by setting and selecting end points based on predetermined conditions, information about the axis may be recorded so that information about the axis can be obtained in processing in step S5, etc., which will be described later. In other words, even if the axis setting function 24c does not record the correspondence of the end points, it can record information about the start point, direction, and distance of the axis as information about the axis.

After the multiple axes are set by the processes of steps S31 to S33, the condition setting function 24d sets conditions related to the closed curve (step S4). Hereinafter, the conditions related to the closed curve are also referred to as closed curve conditions. Hereinafter, examples of the closed curve conditions include the perimeter of the closed curve, the area of the area surrounded by the closed curve, the circularity of the closed curve, and the diameter of a sphere inscribed in the closed curve. In addition, the condition setting function 24d may set conditions for a two-dimensional closed curve obtained by projecting a three-dimensional closed curve onto a plane from a specific direction as the closed curve condition. For example, the condition setting function 24d may set the perimeter of the projected two-dimensional closed curve as the closed curve condition. In addition, the condition setting function 24d may set various conditions related to the form information (shape, area, etc.) of the closed curve as the closed curve condition. The condition setting function 24d sets one or more conditions among these items.

The closed curve condition may be a condition that basically determines a single closed curve that satisfies the condition, or a condition where there are multiple closed curves that satisfy the condition. Examples of conditions that determine a single closed curve include a maximum value, a minimum value, or a value closest to an arbitrary value. Examples of conditions that cause multiple closed curves to satisfy the condition include a condition that is greater than or equal to an arbitrary value, or less than or equal to an arbitrary value.

For example, the output function 24f displays the UI shown in FIG. 6A on the display 22. The condition setting function 24d then accepts an input operation from the user to set the closed curve condition. Specifically, the user selects one of the items such as "perimeter", "area" and "circularity". The user also selects one of the condition contents such as "maximum", "minimum" and "closest to XX". When selecting the condition "closest to XX", the user can also input an arbitrary value "XX". For example, in the case shown in FIG. 6A, the condition setting function 24d sets the condition "area is smallest" as the closed curve condition. According to this condition, it is possible to determine one closed curve that satisfies the condition. That is, in the case shown in FIG. 6A, the condition setting function 24d displays multiple items and contents of the closed curve condition, accepts selection from the user for the items and contents, and sets the closed curve condition based on the accepted items and contents. Also, FIG. 6A is a diagram for explaining the setting of the closed curve condition according to the first embodiment.

As another example, the output function 24f displays a UI as shown in FIG. 6B on the display 22, and the condition setting function 24d accepts an input operation from the user to set the closed curve condition. Specifically, the user selects an arbitrary item from items such as "perimeter", "area" and "circularity". Here, the user may select multiple items. The user also inputs conditions such as a threshold value for the selected item. For example, in the case shown in FIG. 6B, the condition setting function 24d sets a condition of "area is 10 ^mm2 or less and circularity is 0.8 or more" as the closed curve condition. Note that there may be multiple closed curves that satisfy the condition. That is, in the case shown in FIG. 6B, the condition setting function 24d displays multiple items of the closed curve condition to accept a selection from the user, and further accepts input of the contents of the accepted items to set the closed curve condition. Also, FIG. 6B is a diagram for explaining the setting of the closed curve condition according to the first embodiment.

The UI shown in Figures 6A and 6B is merely an example, and various modifications are possible. For example, it may be possible to input various logical expressions (AND, OR, NOT, etc.) to set more complex conditions. In addition, the condition setting function 24d may automatically set a predetermined condition as a closed curve condition.

Next, the identification function 24e identifies a closed curve based on the multiple axes set in step S3 based on the conditions set in step S4 (step S5). For example, the identification function 24e identifies a closed curve that passes through all of the multiple axes set in step S3 based on the conditions set in step S4. In other words, the identification function 24e identifies a closed curve that satisfies the conditions set in step S4 and intersects with all of the multiple axes set in step S3.

For example, the identification function 24e sets one arbitrary point on each of the multiple axes set in step S3. Hereinafter, the arbitrary point set on an axis is referred to as an evaluation point. Next, the identification function 24e obtains a closed curve by connecting the evaluation points on each adjacent axis at the shortest distance along the shape of the mitral valve region R1. Hereinafter, the closed curve obtained by connecting the evaluation points is referred to as a closed curve candidate. Note that, in order to reduce the calculation load, the identification function 24e may treat the line segment connecting adjacent evaluation points as a straight line.

Next, the specifying function 24e calculates the values of the items related to the closed curve condition set in step 4 for the closed curve candidate acquired based on the evaluation points. For example, if the condition "area is minimum" is set as shown in FIG. 6A, the specifying function 24e calculates the area of the closed curve candidate. Also, for example, if the condition "area is ¹⁰ mm2 or less and circularity is 0.8 or more" is set as shown in FIG. 6B, the specifying function 24e calculates the area and circularity of the closed curve candidate. Also, the specifying function 24e stores the calculated values in the memory 21.

Next, the identifying function 24e changes the position of the evaluation point. That is, the identifying function 24e resets the evaluation point set on a certain axis to another position on the axis to which the evaluation point belongs. The identifying function 24e may change the position of the evaluation point on all of the multiple axes set in step S3, or may change the position of the evaluation point on only some of the multiple axes. There is no particular limitation on the method of determining the position of the evaluation point on each axis to be reset. However, for the sake of efficient calculation, it is preferable to control so that the same combination of the evaluation point positions on each axis as the combination of the evaluation point positions on each axis for which the values of the items related to the closed curve condition are not reset.

For example, in the case where multiple axes are set on the mitral valve region as shown in FIG. 7A, the identification function 24e sets multiple measurement points on each axis as shown in FIG. 7B. For example, the measurement points are set at regular intervals on each axis. Then, the identification function 24e selects one of the multiple measurement points for each axis as an evaluation point. Note that in FIG. 7A and FIG. 7B, the multiple axes set by the axis setting function 24c are shown with solid lines, and the tip shape D1 and the annulus shape D2 are shown with dashed lines. Also, in FIG. 7B, the measurement points located on the anterior leaflet of the mitral valve are shown with white rectangular dots, and the measurement points located on the posterior leaflet are shown with white triangular dots. Also, FIG. 7A and FIG. 7B are diagrams for explaining the setting of evaluation points according to the first embodiment.

For example, the identification function 24e selects one of a plurality of measurement points as an evaluation point for each axis. Next, the identification function 24e acquires a closed curve candidate by connecting the evaluation points, and calculates values of items related to the closed curve condition for the closed curve candidate. Next, the identification function 24e changes the measurement points to be selected for some or all of the plurality of axes, resets the evaluation points, acquires a closed curve candidate connecting the reset evaluation points, and calculates values of items related to the closed curve condition for the closed curve candidate. Similarly, the identification function 24e repeatedly resets the evaluation points and calculates values of items related to the closed curve condition.

That is, the specific function 24e sets an evaluation point on each of the multiple axes, and obtains closed curve candidates by connecting the evaluation points while repeatedly resetting the evaluation points, thereby obtaining multiple closed curve candidates. In addition, the specific function 24e calculates values of items related to the closed curve conditions for each of the multiple closed curve candidates.

The specific function 24e may process all combinations of measurement points, or only some combinations. For example, the specific function 24e may repeat resetting of evaluation points until values of items related to closed curve conditions are calculated for all combinations of measurement points in the range from end point to end point on each axis. Alternatively, the specific function 24e may control so that evaluation points are reset only for a predetermined range. For example, the specific function 24e may control so that only measurement points within a certain distance from the end points are reset as evaluation points. Furthermore, the specific function 24e may control so that resetting of evaluation points is terminated when a certain number of times or processing time has been repeated.

Then, the identification function 24e identifies a closed curve that satisfies the closed curve condition based on the values of the items related to the closed curve condition. For example, if the condition "area is smallest" is set as shown in FIG. 6A, the identification function 24e obtains multiple closed curve candidates by repeatedly resetting the evaluation points, and calculates the area for each of the multiple closed curve candidates. The identification function 24e then identifies the closed curve with the smallest area among the multiple closed curve candidates as the closed curve that satisfies the closed curve condition.

Also, for example, when the condition "area is ¹⁰ mm2 or less and circularity is 0.8 or more" is set as shown in Fig. 6A, the identifying function 24e acquires multiple closed curve candidates by repeatedly resetting the evaluation points, and calculates the area and circularity for each of the multiple closed curve candidates. Then, the identifying function 24e identifies a closed curve that satisfies the condition "area is 10 ^mm2 or less and circularity is 0.8 or more" from among the multiple closed curve candidates. Here, the identifying function 24e may identify multiple closed curves that satisfy the closed curve condition.

The identifying function 24e may perform a comparison each time the evaluation points are reset. In other words, the identifying function 24e may record information of the closed curve candidate that calculates the value closest to the closed curve condition (such as the position of each evaluation point and the value of the item related to the closed curve condition), compare it with the value in the reset closed curve candidate, and only if the value in the reset closed curve candidate is closer, record the information in the reset closed curve candidate, and delete (overwrite and save) the information that was compared. For example, if a closed curve condition of "area is the smallest" is set, the area of a certain closed curve candidate is X1, and the area of the reset closed curve candidate is X2, the identifying function 24e may perform overwriting and saving if "X1>X2", and discard the information in the reset closed curve candidate if "X1≦X2".

The identifying function 24e may also obtain multiple closed curve candidates by repeatedly resetting the evaluation point according to the value of the item related to the closed curve condition calculated for the closed curve candidate. For example, when the value after moving the evaluation point on a certain axis in a certain direction is farther from the closed curve condition than the value before the change, the identifying function 24e may end the search in that direction on the axis. In other words, the identifying function 24e may obtain multiple closed curve candidates by moving the evaluation point so that the value of the item related to the closed curve condition satisfies or approaches the closed curve condition.

For example, when a closed curve condition of "area is smallest" is set, if the area of a closed curve candidate is calculated and then the area of the closed curve candidate after moving an evaluation point on a certain axis in a certain direction becomes large, the identification function 24e ends the movement of the evaluation point in that direction. This allows the identification function 24e to reduce the amount of calculation and to speed up the search process. However, in this method, the result depends on the initial position of each evaluation point, so it is preferable to set the initial position to a position where the measurement value is best. For example, when a closed curve condition of "area is smallest" is set and the mitral valve is the target organ, the identification function 24e sets the initial position of each evaluation point near the valve orifice or at a position where the axis intersects with a cross section with the smallest area among cross sections perpendicular to the core line. The identification function 24e may also set multiple initial positions for each evaluation point. For example, the identification function 24e performs a similar search at multiple initial positions and obtains a combination of evaluation points with the best measurement value among the searched results using each initial position. For example, the identification function 24e identifies a closed curve with a smallest area for each of a plurality of initial positions, and further identifies the closed curve with the smallest area from among the identified plurality of closed curves.

Although the case where step S5 is executed after step S4 has been described, the processing order in this respect can be changed. For example, the identification function 24e sets an evaluation point for each of the multiple axes set by the axis setting function 24c, and obtains multiple closed curve candidates based on the evaluation points. Next, the condition setting function 24d sets conditions related to the closed curve, for example by accepting an operation from the user. Then, the identification function 24e identifies a closed curve based on the multiple axes based on the closed curve conditions. In other words, the processing circuit 24 may calculate multiple closed curve candidates in advance before the user specifies the closed curve conditions, etc.

Here, the output function 24f may display the closed curve identified in step S5 or the area surrounded by the closed curve. For example, the output function 24f may display the measurement points as shown in FIG. 7B, and may further display the evaluation points that constitute the identified closed curve. In FIG. 7B, the evaluation points that constitute the identified closed curve are shown as round black dots. For example, if the closed curve condition is "minimum area," the round dots in FIG. 7B indicate the closed curve with the smallest area in the mitral valve. Alternatively, the output function 24f may omit displaying the measurement points and display only the evaluation points that constitute the identified closed curve.

In addition to the display in FIG. 7B, the output function 24f may also display at a different angle, for example, as shown in FIG. 8A. Note that while FIG. 7B is a view of the mitral valve as seen from above, FIG. 8A is a view of the mitral valve as seen from the side. In FIG. 8A, the evaluation points that make up the identified closed curve are shown as round black dots. FIG. 8A is a diagram showing a display example according to the first embodiment.

Although Fig. 7B and Fig. 8A have described the case where three-dimensional data showing the distribution of evaluation points and measurement points constituting the identified closed curve is displayed, the embodiment is not limited to this. For example, the output function 24f may select an arbitrary cross section from the three-dimensional data and display it as two-dimensional data, as shown in Fig. 8B. Fig. 8B is a diagram showing a display example according to the first embodiment. In Fig. 8B, the evaluation points constituting the identified closed curve are shown as round black dots, the measurement points located at the anterior leaflet of the mitral valve are shown as square white dots, and the measurement points located at the posterior leaflet are shown as triangular white dots.

That is, the output function 24f may display the distribution of the evaluation points and measurement points that make up the identified closed curve on an arbitrary cross section. The position and angle of the arbitrary cross section may be set in advance, or may be changed by the user. When the arbitrary cross section intersects with the identified closed curve, the output function 24f may display the two evaluation points that make up the identified closed curve, for example, as shown in FIG. 8B.

The output function 24f may also superimpose and display the measurement points and the evaluation points constituting the identified closed curve on the medical image, as shown in FIG. 8B, for example. For example, the output function 24f generates a two-dimensional image of an arbitrary cross section passing through the mitral valve from a three-dimensional image including the mitral valve, and superimposes and displays the measurement points and evaluation points on each position on the two-dimensional image. The output function 24f may also superimpose and display only the measurement points on the medical image, as shown in FIG. 8C, for example. FIG. 8C is a diagram showing a display example according to the first embodiment. The output function 24f may also superimpose and display only the evaluation points constituting the identified closed curve on the medical image.

The output function 24f may also project and display the identified closed curve on an arbitrary cross section. For example, the output function 24f first generates a two-dimensional image of an arbitrary cross section passing through the mitral valve from a three-dimensional image including the mitral valve. The output function 24f also causes the two-dimensional image to be superimposed on the measurement points located on the arbitrary cross section in association with the measurement points. Furthermore, the output function 24f projects and displays the evaluation points constituting the identified closed curve on the two-dimensional image. That is, regardless of whether each evaluation point constituting the identified closed curve is located on the arbitrary cross section, the output function 24f projects in a direction perpendicular to the arbitrary cross section and superimposes each evaluation point on the two-dimensional image. In this way, the output function 24f can display the closed curve identified in the three-dimensional space on the two-dimensional image. Note that FIG. 8D is a diagram showing a display example according to the first embodiment.

The output function 24f may switch between various displays such as FIG. 7B, FIG. 8A, FIG. 8B, FIG. 8C, and FIG. 8D in response to an instruction from a user. In addition, although the case where the identified closed curve is displayed has been described, the output function 24f may also display an area surrounded by the identified closed curve. For example, the output function 24f may display a plane or a curved surface surrounded by the identified closed curve instead of the evaluation points that constitute the identified closed curve.

That is, the output function 24f displays the closed curve or area identified by the identification function 24e. For example, as shown in FIG. 8B, the output function 24f may display the closed curve or area identified by the identification function 24e in association with a medical image. Also, for example, as shown in FIG. 7B, the output function 24f may display the closed curve or area identified by the identification function 24e in association with multiple axes or measurement points set by the axis setting function 24c. Here, the output function 24f may display the evaluation points and measurement points that make up the identified closed curve in different display forms, as shown in FIG. 7B.

By displaying the closed curve identified in step S5, the medical image processing device 20 can provide information about the mitral valve. For example, if a closed curve is identified under the closed curve condition of "minimum area," the user can understand the shape and size of the bottleneck in the mitral valve by referring to the display of the closed curve.

The output function 24f may also calculate measurements of the closed curve identified in step 5 or the area surrounded by the closed curve (step S6) and display them (step S7). For example, the output function 24f may calculate measurements such as the perimeter of the identified closed curve, the area of the area surrounded by the closed curve, and the circularity of the closed curve. The output function 24f may also calculate measurements of a two-dimensional closed curve obtained by projecting the identified closed curve in any direction. The output function 24f may also calculate measurements after approximating the identified closed curve to another shape. For example, the output function 24f may approximate the identified closed curve to an ellipse and calculate measurements such as the major axis and minor axis of the ellipse.

Note that some of the measurement values related to the closed curve may have already been calculated in step S5. For example, if the closed curve condition is "minimum area," the identification function 24e calculates the area of the closed curve in the process of identifying a closed curve that satisfies the closed curve condition. The identification function 24e records the measurement values calculated in step S5 in the memory 21, and the output function 24f may use the measurement values recorded in the memory 21 to perform the display in step S7.

For example, the output function 24f displays a list of the measurement values calculated for multiple items on the display 22. When displaying the closed curve identified in step S5, the output function 24f may display the measurement values alongside or overlaid on the closed curve. For example, as shown in FIG. 9, the output function 24f displays the measurement points and the evaluation points that make up the identified closed curve in association with each other and overlaid on the medical image, and further displays the measurement values calculated for measurement items such as "area," "perimeter," and "circularity" overlaid on the medical image. Note that FIG. 9 is a diagram showing a display example according to the first embodiment.

By calculating and displaying the measurement values, the medical image processing device 20 can provide information about the mitral valve. For example, if a closed curve is identified under the closed curve condition of "minimum area," the user can refer to the display of the measurement values related to the closed curve and grasp the area, circumference, circularity, and other numerical values of the bottleneck portion of the mitral valve.

In addition, various modifications of the display form are possible. For example, the output function 24f may change the display form (color, shape, etc.) of the displayed closed curve or the measurement value based on the magnitude of the measurement value. The output function 24f may also change the display form based on the relationship between the cardiac phase at which the measurement value was calculated and the magnitude of the measurement value, or may display a screen that calls attention. For example, if the closed curve is specified as the closed curve condition of "minimum area," and the area of the closed curve is large during systole or small during diastole, the presence of a disease related to the mitral valve is suspected. In such a case, the output function 24f may display a warning or highlight the measurement value such as the area.

The output function 24f may also display the identified closed curve or area, and may also display a cross section at a different angle from the closed curve or area. For example, the output function 24f may generate and display a cross section image that is approximately parallel to the identified closed curve from a three-dimensional medical image, and may also display the identified closed curve on the cross section image. Furthermore, the output function 24f may generate and display, for example, a perpendicular cross section that is perpendicular to the cross section image.

For example, the output function 24f may further display various analysis images. For example, the output function 24f performs fluid analysis based on multiple medical images collected in time series, and calculates fluid properties such as the flow direction and flow velocity of blood flow near the mitral valve. The output function 24f also displays the calculated fluid properties, for example, in a color map. Furthermore, the output function 24f can also display a closed curve or area identified by the identification function 24e on such a color map.

So far, the mitral valve has been described as the target organ, but the present invention can be applied to valves other than the mitral valve in the same way. In addition, the present invention can be applied to various biological organs that have a tubular structure, such as blood vessels, esophagus, trachea, ureter, and intestine, in addition to valves. In other words, the above-mentioned embodiment can be applied to any biological organ through which fluids such as blood or gas, or solids, flow.

As an example, the acquisition function 24b acquires a medical image including blood vessels such as coronary arteries and cerebral arteries. The axis setting function 24c sets multiple axes based on the blood vessels. For example, the axis setting function 24c sets a first closed curve at one end of the blood vessel included in the medical image, sets a second closed curve at the other end, and sets multiple end points on each of the first closed curve and the second closed curve. The axis setting function 24c then sets multiple axes based on the multiple end points that have been set. For example, the axis setting function 24c sets multiple axes by connecting the end points along the blood vessel shape.

Here, blood vessels may have a stenosis. For example, as shown in FIG. 10, stenosis may occur in blood vessels such as coronary arteries and cerebral arteries due to intravascular structures such as plaque H1 and plaque H2. In such cases, the axis setting function 24c sets multiple axes along the shape of the blood vessel and the shape of the intravascular structure. That is, the axis setting function 24c sets multiple axes by connecting the end points set on the first closed curve and the end points set on the second closed curve along the blood vessel or the intravascular structure. Note that FIG. 10 is a diagram showing an example of an intravascular stenosis according to the first embodiment.

The condition setting function 24d also sets a closed curve condition for the closed curve corresponding to the blood vessel. The identification function 24e also identifies a closed curve based on the multiple axes set based on the closed curve condition. For example, if the closed curve condition is "area is minimum", the identification function 24e identifies a closed curve that passes through all the multiple axes set and has the minimum area. Here, the closed curve that passes through all the multiple axes is set at multiple positions in the blood vessel as shown in FIG. 10, for example. For example, the closed curve that passes through all the number axes is set in a region with diameter G1 as the major axis, a region with diameter G2 as the major axis, a region with diameter G3 as the major axis, and the like. Here, the identification function 24e can identify the closed curve set in the region with diameter G3 as the major axis as the closed curve with the minimum area. That is, the identification function 24e can identify the region with diameter G3 as the major axis as the bottleneck region.

As shown in FIG. 10, the shape inside a blood vessel may be complicated due to the presence of plaque, etc. Here, for example, when evaluating the orthogonal cross-sectional area at each position along the center line of a blood vessel, there is a risk that a region with a major axis of diameter G1 or diameter G2 may be erroneously determined to be a bottleneck region. In response to this, the medical image processing device 20 can set multiple axes to evaluate closed curves that are inclined with respect to the center line of the blood vessel, such as a region with a major axis of diameter G3, and appropriately identify the bottleneck region.

As described above, according to the first embodiment, the acquisition function 24b acquires a medical image including a biological organ or valve having a tubular structure. The axis setting function 24c sets multiple axes based on the biological organ or valve having a tubular structure. The condition setting function 24d sets a closed curve condition for a closed curve corresponding to the biological organ or valve having a tubular structure. The identification function 24e identifies a closed curve based on the multiple axes set by the axis setting function 24c or an area surrounded by the closed curve based on the closed curve condition. Therefore, the medical image processing device 20 according to the first embodiment can provide information about a biological organ or valve having a tubular structure.

For example, in the case of the mitral valve, if the area of the tip of the valve is the smallest, it is easy to identify the bottleneck, but mitral valves come in a variety of shapes, and the area between the tip and the annulus may be the bottleneck. That is, the closed curve with the smallest area may be located on the surface of the valve leaflet, rather than at the tip. Alternatively, the closed curve with the smallest area may have part passing through the tip of the valve leaflet, and other parts located on the surface of the valve leaflet. Thus, while it is possible to obtain shape information of the mitral valve based on medical images, it is not easy to obtain information such as the area of the bottleneck. In response to this, the medical image processing device 20 can appropriately identify closed curves even for organs with complex shapes, and provide that information.

The mitral valve is composed of two leaflets, an anterior leaflet and a posterior leaflet. In this way, biological organs such as blood vessels and valves can sometimes be divided into multiple regions. The medical image processing device 20 may process the biological organ as a single region, or may process characteristic regions or regions with different characteristics separately from the region of interest.

For example, in step S2 of FIG. 2, the axis setting function 24c may divide a biological organ such as a blood vessel or a valve into multiple regions and acquire a region of interest for each region. For example, the axis setting function 24c acquires two leaflets, the anterior leaflet and the posterior leaflet of the mitral valve, as separate regions. That is, the axis setting function 24c performs division based on the anatomical characteristics or structure of the biological organ such as a blood vessel or a valve. Here, the output function 24f may display the acquired multiple regions of interest on the display 22. The output function 24f may also control the multiple regions of interest to be displayed under different display conditions, such as changing the color of the region of interest based on the anterior leaflet and the region of interest based on the posterior leaflet. The output function 24f may also control the display of only a part of the multiple regions of interest in response to a user instruction.

Also, for example, in step S3, the axis setting function 24c may divide a biological organ such as a blood vessel or a valve into multiple regions and set multiple axes using different criteria for each region. That is, the axis setting function 24c may set the number of axes to be set and the intervals, etc. for each region such as the anterior leaflet and the posterior leaflet. For example, the axis setting function 24c can perform control so that six axes are set for the anterior leaflet and four axes are set for the posterior leaflet.

Also, for example, in step S4, the condition setting function 24d may divide a biological organ such as a blood vessel or a valve into a plurality of regions and set a closed curve condition for each region. For example, the condition setting function 24d may curve-approximate each line segment and set the curvature. Alternatively, for example, the condition setting function 24d may divide the region surrounded by the closed curve into two regions, an anterior cusp side region and a posterior cusp side region, and set the area of each region as the closed curve condition.

Also, for example, in step S5, the identifying function 24e may divide a biological organ such as a blood vessel or a valve into a plurality of regions, and identify a closed curve or a region surrounded by the closed curve using different criteria for each region. For example, the identifying function 24e may set different conditions for determining the position at which the evaluation point is reset for the evaluation point on the axis on the anterior cusp side and the evaluation point on the axis on the posterior cusp side. For example, the identifying function 24e may control so that only the posterior cusp side is reset, without resetting the anterior cusp side. Also, for example, when the repeated processing is stopped after a certain number of times, the identifying function 24e may control so that the number of times the posterior cusp side is reset is greater than the number of times the anterior cusp side is reset.

Also, for example, in steps S6 and S7, the output function 24f may divide biological organs such as blood vessels and valves into multiple regions and calculate the measurement value for each region. For example, as shown in FIG. 11, the output function 24f may calculate and display each of the anterior cusp circumference and the posterior cusp circumference. The output function 24f may display the closed curve or region specified by the specification function 24e in a different display mode for each region. For example, as shown in FIG. 11, the output function 24f may display the closed curve so that the evaluation point on the axis on the anterior cusp side and the evaluation point on the axis on the posterior cusp side have different shapes. In FIG. 11, the evaluation point on the anterior cusp side of the specified closed curve is indicated by a black square dot, and the evaluation point on the posterior cusp side is indicated by a black triangular dot. FIG. 11 is a diagram showing a display example according to the first embodiment.

The method of dividing biological organs such as blood vessels and valves into multiple regions for processing may be applied to only one of steps S2 to S7 in FIG. 2, or to multiple or all of them.

In the above-described embodiment, a case where multiple axes are set according to the flowchart in FIG. 4A has been described. That is, a first closed curve is set at one end of the mitral valve or the like, a second closed curve is set at the other end, multiple end points are set on each of the first closed curve and the second closed curve, and multiple axes are set based on the multiple end points. However, the embodiment is not limited to this.

For example, the axis setting function 24c may set multiple axes using multiple predetermined cross sections. For example, the axis setting function 24c sets multiple axes using a UI shown in FIG. 12. Note that FIG. 12 shows a case where multiple predetermined cross sections are defined radially. That is, the axis setting function 24c defines radial cross sections such as cross sections P1 to P4, and sets multiple axes based on the positions at which the cross sections and the region of interest intersect. Note that the cross sections P1 to P4 are two-dimensional planes having a length in the depth direction. The center position of the radial cross sections may be set manually by the user or based on a predetermined criterion. For example, the center position of the radial cross section may be the center of the display screen, or may be the center of gravity of the region of interest or the center of gravity of the tip closed curve or the annulus closed curve. However, since it is desirable to set the center position of the radial cross section inside the region of interest, the axis setting function 24c may be controlled so that it cannot be set outside the region of interest. Note that FIG. 12 is a diagram showing an example of an axis setting method according to the first embodiment.

The axis setting function 24c also determines the direction in which the cross section is set. For example, the direction in which the cross section is set may be from the front side to the back side of the display screen, or may be from the center of gravity or center of the distal end closed curve toward the center of gravity or center of the annulus closed curve.

The axis setting function 24c also sets information regarding the number of cross sections to be set. The number of cross sections may be determined in advance, or may be set appropriately using, for example, the UI of FIG. 12. FIG. 12 shows a case where four cross sections, P1 to P4, are set as the number of cross sections. Alternatively, instead of the number of cross sections, the angle between the cross sections may be set. The axis setting function 24c then sets the set number of cross sections, or the number of cross sections that can be set at equal intervals and closest to the set angle between the cross sections. That is, when the angle between the cross sections is set, the number of cross sections is determined by dividing "180 degrees" by the set angle and discarding the decimal point. For example, when "45 degrees" is set as the angle between the cross sections, four cross sections are set as shown in FIG. 12. That is, in the case shown in FIG. 12, the axis setting function 24c can define a predetermined number of cross sections by receiving input of the number of cross sections or the angle between the cross sections.

The axis setting function 24c may also perform control so that after a cross section is set, the user can move, delete, or add the cross section. For example, the output function 24f displays the set cross section on the display 22, and the axis setting function 24c accepts user operations on the cross section based on, for example, the operation of the mouse pointer M. The axis setting function 24c may also perform control so that movement beyond adjacent cross sections is not possible.

After the positions and number of radial cross sections are determined, the axis setting function 24c sets multiple axes based on the intersection positions of these cross sections and the region of interest. That is, the axis setting function 24c sets multiple axes using multiple cross sections defined radially. For example, the axis setting function 24c sets multiple axes based on the intersection points of each cross section and the tip closed curve or the annulus closed curve as endpoints. For example, the axis setting function 24c sets multiple axes by connecting the endpoints of the tip closed curve and the endpoints of the annulus closed curve along the shape of the region of interest.

In addition, in Figures 6A and 6B, a case has been described in which the closed curve conditions are set by the user directly inputting the closed curve conditions. For example, in Figure 6A, it has been described that the closed curve conditions are selected from "perimeter," "area," "circularity," and the like. However, there are cases in which the user does not understand the correspondence between these closed curve conditions and their clinical meanings.

Therefore, the condition setting function 24d may set the closed curve condition by presenting the clinical meaning of each closed curve condition to the user and accepting the selection of the clinical meaning from the user. For example, as shown in FIG. 13A, the output function 24f displays the clinical meaning of the smallest area "the size of the area that becomes a bottleneck of blood flow", the clinical meaning of the largest circularity "the ease of blood flow", and the clinical meaning of the largest perimeter "the degree of spread of the valve orifice", etc., in a table. Also, for example, as shown in FIG. 13B, the output function 24f displays multiple clinical meanings in a pull-down menu based on a user operation received via the mouse pointer M. Then, the user selects one of the multiple clinical meanings that is of interest, and the condition setting function 24d sets the closed curve condition corresponding to the selected clinical meaning. That is, in the case shown in FIG. 13A, the output function 24f displays multiple combinations of clinical meanings and the above conditions, and the condition setting function 24d can set the closed curve condition based on the user operation for the multiple combinations displayed. In the case shown in FIG. 13B, the output function 24f displays multiple clinical meanings in a pull-down menu, and the condition setting function 24d can set the closed curve conditions based on a user operation on the pull-down menu. Note that FIG. 13A and FIG. 13B are diagrams showing an example of a method for setting the closed curve conditions according to the first embodiment.

Second Embodiment
In the second embodiment, a case where a closed curve or an area surrounded by the closed curve is specified based on a plurality of medical images acquired in a time series will be described. That is, in the second embodiment, a case where a closed curve or an area surrounded by the closed curve is specified based on information of a plurality of images captured at different times will be described. The medical image processing system 1 according to the second embodiment has a configuration similar to that of the medical image processing system 1 shown in FIG. 1, and is different in part of the processing by the axis setting function 24c, the condition setting function 24d, the specifying function 24e, and the output function 24f. Hereinafter, the points described in the above embodiment are assigned the same reference numerals as those in FIG. 1, and the description will be omitted.

For example, in step S1 shown in FIG. 2, the acquisition function 24b acquires multiple medical images in a time series. For example, the acquisition function 24b acquires multiple medical images including the mitral valve, with different cardiac phases. In addition, in step S2, the axis setting function 24c acquires a region of interest for each of the multiple medical images. In addition, in step S3, the axis setting function 24c sets multiple axes for each of the multiple medical images. Note that it is desirable to set the same conditions for each medical image, such as the number of axes.

In addition, in step S4, the condition setting function 24d sets a condition based on multiple medical images as a closed curve condition. Examples of such conditions include "the amount of variation of the closed curve between medical images is minimal (or maximal)", "the amount of change in area between medical images is minimal (or maximal)", "the average value of the area in multiple medical images is minimal (or maximal)", etc.

In step S5, the identification function 24e also associates the evaluation points on each axis between medical images. For example, the identification function 24e sets the same number of measurement points under the same conditions in all medical images, and associates the corresponding measurement points between medical images. In another example, the identification function 24e may set measurement points in a reference medical image, and perform alignment between the reference medical image and other medical images, so that positions corresponding to the measurement points in the reference medical image are set as measurement points in the other medical images. Note that since the shape of the mitral valve changes for each medical image, the distance and positional relationship between the measurement points will change for each medical image. Then, the identification function 24e calculates values of items related to the closed curve conditions based on multiple medical images based on the corresponding closed curves in each medical image. The identification function 24e also identifies the closed curve or the area surrounded by the closed curve by repeating the process of resetting the evaluation points and calculating the values of items related to the closed curve conditions.

The output function 24f also calculates the measurement value in step S6 and displays the measurement value in step S7. Here, the measurement value calculated by the output function 24f may be calculated based on multiple medical images, such as the "amount of variation of a closed curve between medical images," or may be calculated based on a single medical image, such as the "area of a closed curve."

Third Embodiment
Biological organs such as blood vessels and valves may include characteristic points or structures. In the third embodiment, a case where processing is performed taking such characteristic points or structures into consideration will be described. A medical image processing system 1 according to the third embodiment has a configuration similar to that of the medical image processing system 1 shown in FIG. 1, and is different in some of the processing by the axis setting function 24c, the condition setting function 24d, the identification function 24e, and the output function 24f. Hereinafter, the points described in the above embodiment are denoted by the same reference numerals as in FIG. 1, and the description will be omitted.

For example, in step S2 shown in FIG. 2, the axis setting function 24c acquires a region of interest and identifies a characteristic point or structure. Here, the characteristic point or structure is, for example, the position of the commissure of the anterior and posterior leaflets of the mitral valve, the position of the fibrous trigone, which is a position related to the mitral valve, or the center position of the apex or aortic valve. For example, the axis setting function 24c can identify the characteristic point or structure by the known region extraction technique described above. The axis setting function 24c may simultaneously acquire the region of interest and identify the characteristic point or structure using the same method, or may separately use different methods.

For example, the axis setting function 24c can use a characteristic point or structure when setting an axis in step S3. That is, the axis setting function 24c can set multiple axes based on the characteristic point or structure. Below, a case where an axis is set according to the flowchart in FIG. 4 will be described. For example, as an example of step S31, a case where the pixel located at the bottom end or top end in each pixel row in a predetermined direction is specified as the tip shape D1 and the annulus shape D2 has been described. Here, the axis setting function 24c can specify the predetermined direction based on the characteristic point or structure. Also, as an example of step S32, a case where a starting point is set based on a reference point and multiple end points are set from the starting point has been described. Here, the axis setting function 24c can set a reference point or a starting point based on a characteristic point or structure.

In addition, the identification function 24e can use a characteristic point or structure when identifying a closed curve or an area surrounded by the closed curve in step S5. That is, the identification function 24e can identify a closed curve or an area surrounded by the closed curve based on the characteristic point or structure. For example, first, in step S4, the condition setting function 24d sets the presence or absence of use of the position of the characteristic point or structure as a closed curve condition. For example, the condition setting function 24d can set the presence or absence of use of the position of the characteristic point or structure by accepting a user operation via a UI such as a checkbox. Then, the identification function 24e identifies a closed curve passing through the characteristic point or structure or an area surrounded by the closed curve based on the closed curve condition. For example, the identification function 24e identifies a closed curve that passes through all axes set by the axis setting function 24c, passes through two commissures, and has a minimum perimeter. In addition, for example, the identification function 24e may set the search range of the evaluation point to be moved on each axis based on the position (coordinates) of the characteristic point or structure when identifying a closed curve. For example, the specific function 24e may be controlled to move the evaluation point only in an area surrounded by four points: two commissures and two fiber triangles.

The output function 24f may also display characteristic points or structures so that the user can recognize them. For example, when displaying a region of interest, an axis, an evaluation point, a closed curve identified by the identification function 24e or a region surrounded by the closed curve, measurement values related to the closed curve or the region, etc., the position of the characteristic points or structures may also be displayed. For example, the output function 24f may display the characteristic points or structures using a display form (color, shape) different from other evaluation points, regions, etc.

Fourth Embodiment
Although the first to third embodiments have been described above, the present invention may be embodied in various different forms other than the above-described embodiments.

For example, in the above-described embodiment, a case has been described in which an area of interest is acquired and axes are set on the area of interest. However, the embodiment is not limited to this. For example, the axis setting function 24c may omit acquiring the area of interest and set multiple axes directly from the medical image acquired by the acquisition function 24b.

For example, the axis setting function 24c can set multiple axes without acquiring a region of interest by machine learning techniques such as deep learning. As an example, the axis setting function 24c first acquires a region of interest from a medical image as described in the first embodiment, and sets the multiple axes based on the region of interest. The axis setting function 24c also collects a large number of combinations of the medical image and the multiple axes that have been set as learning data. The axis setting function 24c also executes machine learning with the medical image as input data and the multiple axes that have been set as output data to generate a trained model. Such a trained model is functionally configured to set multiple axes upon receiving an input of a medical image. Then, when the acquisition function 24b acquires a new medical image, the axis setting function 24c can set multiple axes without acquiring a region of interest by inputting the medical image into the trained model. That is, step S2 shown in FIG. 2 can be omitted. Similarly, the above-mentioned measurement points can also be set without acquiring a region of interest.

For example, in the above embodiment, a case where multiple axes are generated each time has been described. However, the embodiment is not limited to this, and the axis setting function 24c may set multiple axes by transforming multiple predetermined axes.

For example, the axis setting function 24c acquires the cylindrical model shown in FIG. 14 in advance and stores it in the memory 21. Multiple axes are set in the cylindrical model based on arbitrary conditions. In other words, the cylindrical model is an example of multiple predetermined axes. In the case shown in FIG. 14, ten linear axes are set in the cylindrical model along the height direction. Note that FIG. 14 is a diagram for explaining the axis setting method according to the fourth embodiment.

For example, when the acquisition function 24b acquires a new medical image, the axis setting function 24c reads out the cylindrical model from the memory 21 and deforms it to match the structure of a biological organ such as a mitral valve. For example, the axis setting function 24c sets multiple axes by aligning the read cylindrical model with a medical image including a biological organ such as a mitral valve. That is, the axis setting function 24c performs deformation based on the alignment result between multiple predetermined axes and a biological organ such as a mitral valve. For example, the axis setting function 24c acquires a region of interest from a medical image and aligns the cylindrical model so as to match the shape of the region of interest. Note that a known deformation alignment method can be used as the alignment method. For example, the axis setting function 24c can align the cylindrical model with the medical image by a method of aligning lattice points such as ICP (Iterative Closest Point). The axis setting function 24c can also set the axis by aligning images using an image in which an axis has already been set. In this case, known methods such as the Free-Form Deformation (FFD) method or the Large Deformation Diffeomorphic Metric Mapping (LDDMM) method can be used to align the images.

In the above-described embodiment, the display 22 is displayed based on the closed curve or area identified by the identification function 24e. However, the embodiment is not limited to this. For example, the output function 24f may transmit the closed curve or area identified by the identification function 24e to an external device. In this case, the external device can display the closed curve or area, and calculate and display various measurement values.

The term "processor" used in the above description refers to circuits such as a CPU, a GPU (Graphics Processing Unit), an Application Specific Integrated Circuit (ASIC), a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)). When the processor is, for example, a CPU, the processor realizes functions by reading and executing a program stored in a memory circuit. On the other hand, if the processor is, for example, an ASIC, instead of storing the program in a memory circuit, the function is directly built into the circuit of the processor as a logic circuit. Note that each processor in the embodiments is not limited to being configured as a single circuit for each processor, and may be configured as a single processor by combining multiple independent circuits to realize the function. Furthermore, multiple components in each figure may be integrated into a single processor to realize the function.

In addition, in FIG. 1, a single memory 21 is described as storing programs corresponding to each processing function of the processing circuit 24. However, the embodiment is not limited to this. For example, a configuration may be adopted in which a plurality of memories 21 are distributed and the processing circuit 24 reads out corresponding programs from the individual memories 21. Also, instead of storing programs in the memory 21, the programs may be directly built into the circuitry of the processor. In this case, the processor realizes the functions by reading and executing the programs built into the circuitry.

The components of each device according to the above-described embodiments are conceptual and functional, and do not necessarily need to be physically configured as shown in the figures. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figures, and all or part of the devices can be functionally or physically distributed and integrated in any unit depending on various loads and usage conditions. Furthermore, all or any part of the processing functions performed by each device can be realized by a CPU and a program analyzed and executed by the CPU, or can be realized as hardware using wired logic.

The medical image processing method described in the above embodiment can be realized by executing a prepared program on a computer such as a personal computer or a workstation. This program can be distributed via a network such as the Internet. This program can also be recorded on a non-transitory recording medium that can be read by a computer, such as a hard disk, flexible disk (FD), CD-ROM, MO, or DVD, and executed by being read from the recording medium by a computer.

According to at least one of the embodiments described above, it is possible to provide information about a biological organ or valve having a tubular structure.

Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are within the scope of the invention and its equivalents as set forth in the claims, as well as the scope and spirit of the invention.

Regarding the above embodiment, the following supplementary notes are disclosed as one aspect and optional features of the invention.
(Appendix 1)
An acquisition unit that acquires medical images including blood vessels or valves;
an axis setting unit that sets a plurality of axes based on the blood vessel or the valve;
a condition setting unit for setting a condition related to a closed curve corresponding to the blood vessel or valve;
and a specification unit that specifies a closed curve based on the multiple axes or an area surrounded by the closed curve based on the condition.
(Appendix 2)
The axis setting unit may obtain a region of interest from the medical image based on the blood vessel or the valve, and set the multiple axes based on the region of interest.
(Appendix 3)
The axis setting unit may divide the blood vessel or the valve into a plurality of regions and obtain the region of interest for each of the regions.
(Appendix 4)
The axis setting unit may set a first closed curve at one end of the blood vessel or valve, set a second closed curve at the other end, set a plurality of end points on each of the first closed curve and the second closed curve, and set the plurality of axes based on the plurality of end points.
(Appendix 5)
The multiple axes may be set along the shape of the blood vessel or the valve.
(Appendix 6)
the medical image includes a blood vessel;
The axis setting unit may set the multiple axes by connecting an end point set on the first closed curve and an end point set on the second closed curve along the blood vessel or an intravascular structure.
(Appendix 7)
The axis setting unit may set different numbers of the end points on the first closed curve and the second closed curve.
(Appendix 8)
The axis setting unit may set the multiple endpoints on each of the first closed curve and the second closed curve by accepting input of a starting point for sequentially setting the multiple endpoints and conditions for setting the multiple endpoints.
(Appendix 9)
The axis setting unit may receive a user operation on a display of the first closed curve and the second closed curve as an input of the starting point.
(Appendix 10)
The axis setting unit may set the multiple axes further based on characteristic points or structures in the blood vessels or valves.
(Appendix 11)
The axis setting section may set the plurality of axes using a plurality of predetermined cross sections.
(Appendix 12)
The predetermined cross sections may be defined radially.
(Appendix 13)
The axis setting unit may define the predetermined plurality of cross sections by receiving an input of a number of cross sections or an angle between the cross sections.
(Appendix 14)
The axis setting unit may set the multiple axes using a trained model that is functionalized to set multiple axes based on input of a medical image.
(Appendix 15)
The axis setting unit may set a plurality of predetermined axes by transforming the plurality of predetermined axes.
(Appendix 16)
The axis setting portion may perform the deformation in accordance with a structure of the blood vessel or the valve.
(Appendix 17)
The axis setting unit may perform the deformation based on a result of alignment between a predetermined axis and the blood vessel or the valve.
(Appendix 18)
The plurality of predetermined axes may be cylindrical models.
(Appendix 19)
The identification unit may acquire a plurality of closed curve candidates that pass through all of the plurality of axes, calculate values of items related to the conditions for each of the acquired plurality of closed curve candidates, and, based on the values, identify a closed curve based on the plurality of axes or an area surrounded by the closed curve.
(Appendix 20)
The identification unit may set an evaluation point on each of the multiple axes and acquire the closed curve candidate by connecting the evaluation points.
(Appendix 21)
The identification unit may acquire the multiple closed curve candidates by repeatedly resetting the evaluation point in accordance with the value.
(Appendix 22)
The specification unit may acquire the multiple closed curve candidates by moving the evaluation point so that the value satisfies the condition or approaches the condition.
(Appendix 23)
The identification unit may identify a closed curve based on the multiple axes or a region surrounded by the closed curve, further based on a characteristic point or structure in the blood vessel or valve.
(Appendix 24)
The acquisition unit acquires a plurality of the medical images in time series,
The condition setting unit sets the condition based on the plurality of medical images,
The specification unit may specify a closed curve based on the multiple axes or a region surrounded by the closed curve, based on the condition.
(Appendix 25)
The apparatus may further include an output unit that performs output based on the closed curve or area identified by the identification unit.
(Appendix 26)
The output unit may display the closed curve or the area identified by the identification unit.
(Appendix 27)
The output unit may display the closed curve or the area identified by the identification unit in association with the medical image.
(Appendix 28)
The output unit may display the closed curve or the area identified by the identification unit in association with the multiple axes or the measurement points.
(Appendix 29)
The output unit may display the evaluation points and the measurement points constituting the closed curve identified by the identification unit in different display forms.
(Appendix 30)
The output unit may divide the blood vessel or the valve into a plurality of regions, and display the closed curve or the region identified by the identification unit in a different display mode for each region.
(Appendix 31)
The output unit may display a measurement value relating to the closed curve or area identified by the identification unit.
(Appendix 32)
The output unit may divide the blood vessel or valve into a plurality of regions, and calculate and display the measurement value for each region.
(Appendix 33)
The axis setting unit may divide the blood vessel or the valve into a plurality of regions, and set the plurality of axes based on different criteria for each region.
(Appendix 34)
The axis setting unit may divide the blood vessel or the valve into a plurality of regions and set the condition for each of the regions.
(Appendix 35)
The identification unit may divide the blood vessel or the valve into a plurality of regions, and identify a closed curve based on the plurality of axes or a region surrounded by the closed curve using a different criterion for each region.
(Appendix 36)
The condition setting unit may receive a selection of a clinical meaning from a user, and set the condition corresponding to the selected clinical meaning.
(Appendix 37)
an output unit that displays a plurality of combinations of the clinical meaning and the condition;
The condition setting unit may set the condition based on a user operation for the combination.
(Appendix 38)
An output unit that displays multiple clinical meanings in a pull-down menu;
The condition setting unit may set the condition based on a user operation on a pull-down display.
(Appendix 39)
The condition setting unit may display a plurality of items and contents of the condition, accept a selection from a user regarding the item and the content, and set the condition based on the accepted item and content.
(Appendix 40)
The condition setting unit may set the condition by displaying a plurality of items of the condition, accepting a selection from a user, and further accepting input of details for the accepted items.
(Appendix 41)
The decomposition may be characterized as being based on anatomical features or structure of the vessel or valve.
(Appendix 42)
The condition setting unit may set, as the condition, at least one of a perimeter of a closed curve, an area of a region enclosed by the closed curve, a circularity of the closed curve, and a diameter of a sphere inscribed in the closed curve.
(Appendix 43)
The medical image processing apparatus according to claim 1 , wherein the specifying unit specifies a closed curve passing through all of the plurality of axes or a region surrounded by the closed curve based on the condition.
(Appendix 44)
An acquisition unit that acquires a medical image including a biological organ or a valve having a tubular structure;
an axis setting unit that sets a plurality of axes based on the biological organ or the valve;
a condition setting unit for setting a condition related to a closed curve corresponding to the biological organ or valve;
and a specification unit that specifies a closed curve based on the multiple axes or an area surrounded by the closed curve based on the condition.
(Appendix 45)
Acquire a medical image including a blood vessel or a valve;
A plurality of axes are set based on the blood vessel or the valve;
Setting a condition regarding a closed curve corresponding to the blood vessel or valve;
and identifying a closed curve based on the plurality of axes or an area surrounded by the closed curve based on the condition.
(Appendix 46)
A program for causing a computer to execute each component of the above medical image processing device.

REFERENCE SIGNS LIST 1 Medical image processing system 10 Medical image diagnostic device 20 Medical image processing device 21 Memory 22 Display 23 Input interface 24 Processing circuit 24a Control function 24b Acquisition function 24c Axis setting function 24d Condition setting function 24e Identification function 24f Output function 30 Image storage device

Claims (18)

An acquisition unit that acquires medical images including blood vessels or valves;
an axis setting unit that sets a plurality of end points for each of one end and the other end of the shape of the blood vessel or the valve, and sets a plurality of axes based on the plurality of end points ;
a condition setting unit for setting a condition related to a closed curve corresponding to the blood vessel or valve;
and a specification unit that specifies a closed curve passing through the plurality of axes or a region surrounded by the closed curve based on the condition. The medical image processing device according to claim 1, wherein the axis setting unit divides the blood vessel or valve into a plurality of regions, obtains a region of interest from the medical image for each region, and sets the plurality of axes based on the region of interest. 3. The medical image processing device according to claim 1, wherein the axis setting unit sets a first closed curve at one end of the blood vessel or valve, sets a second closed curve at the other end, sets the multiple end points for each of the first closed curve and the second closed curve, and sets the multiple axes based on the multiple end points. The medical image processing device according to claim 3, wherein the multiple axes are set along the shape of the blood vessel or valve. the medical image includes a blood vessel;
5. The medical image processing device according to claim 3, wherein the axis setting unit sets the multiple axes by connecting an end point set on the first closed curve and an end point set on the second closed curve along the blood vessel or an intravascular structure. The medical image processing device according to any one of claims 1 to 5, wherein the axis setting unit sets the multiple axes further based on characteristic points or structures in the blood vessel or valve. The medical image processing device according to claim 1, wherein the axis setting unit sets the multiple axes using multiple predetermined cross sections. The medical image processing device according to any one of claims 1 to 7, wherein the identification unit acquires a plurality of closed curve candidates passing through all of the plurality of axes, calculates values of items related to the condition for each of the acquired plurality of closed curve candidates, and identifies a closed curve passing through the plurality of axes or an area surrounded by the closed curve based on the values. The medical image processing apparatus according to claim 8 , wherein the specifying unit sets an evaluation point on each of the plurality of axes, and acquires the closed curve candidate by connecting the evaluation points. The acquisition unit acquires a plurality of the medical images in time series,
The condition setting unit sets the condition based on the plurality of medical images,
The medical image processing device according to claim 1 , wherein the specifying unit specifies a closed curve passing through the plurality of axes or a region surrounded by the closed curve based on the condition. The medical image processing apparatus according to claim 1 , further comprising an output unit that performs output based on the closed curve or area specified by the specification unit. The medical image processing apparatus according to claim 1 , further comprising an output unit that displays the closed curve or area identified by the identification unit. The medical image processing device according to any one of claims 1 to 12, wherein the condition setting unit sets, as the condition, at least one of a perimeter of a closed curve, an area of a region surrounded by the closed curve, a circularity of the closed curve, and a diameter of a sphere inscribed in the closed curve. The medical image processing device according to claim 1 , wherein the specifying unit specifies a closed curve passing through all of the multiple axes or a region surrounded by the closed curve based on the condition. An acquisition unit that acquires a medical image including a biological organ or a valve having a tubular structure;
an axis setting unit that sets a plurality of end points for each of one end and the other end of the shape of the biological organ or the valve, and sets a plurality of axes based on the plurality of end points ;
a condition setting unit for setting a condition related to a closed curve corresponding to the biological organ or valve;
and a specification unit that specifies a closed curve passing through the plurality of axes or a region surrounded by the closed curve based on the condition. Acquire a medical image including a blood vessel or a valve;
A plurality of end points are set for each of one end and the other end of the shape of the blood vessel or the valve, and a plurality of axes are set based on the plurality of end points ;
Setting a condition regarding a closed curve corresponding to the blood vessel or valve;
and identifying a closed curve passing through the plurality of axes or an area surrounded by the closed curve based on the condition. Acquire a medical image including a blood vessel or a valve;
A plurality of end points are set for each of one end and the other end of the shape of the blood vessel or the valve, and a plurality of axes are set based on the plurality of end points ;
Setting a condition regarding a closed curve corresponding to the blood vessel or valve;
A program that causes a computer to execute each process of identifying a closed curve passing through the multiple axes or an area surrounded by the closed curve based on the condition. An acquisition unit that acquires medical images including blood vessels or valves;
an axis setting unit that sets a plurality of end points for each of one end and the other end of the shape of the blood vessel or the valve, and sets a plurality of deformed axes based on the plurality of end points, the plurality of deformed axes being deformed based on the shape of the blood vessel or the valve;
a condition setting unit for setting a condition related to a closed curve corresponding to the blood vessel or valve;
a specification unit that specifies a closed curve passing through the plurality of deformation axes or a region surrounded by the closed curve based on the condition.

Images