Medical processing apparatus and medical processing method

Abstract

To enable easy extraction of a standard cross-section of a subject from three-dimensional volume data.SOLUTION: A medical processing apparatus according to an embodiment comprises: an acquisition unit that acquires three-dimensional volume data of a subject; a first cross-section extraction unit that extracts a first cross-section intersecting a reference line of the subject included in the three-dimensional volume data; a second cross-section extraction unit that extracts a second cross-section intersecting the first cross-section from the three-dimensional volume data based on a posture of the subject included in the first cross-section; and an information output unit that outputs information based on the second cross-section.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] SUMMARY OF THE INVENTION The embodiments disclosed in this specification and the drawings relate to medical processing devices and methods. [Background technology]

[0002] In fetal examinations, in order to measure fetal crown-rump length (CRL), fetal nuchal translucency (NT) area, fetal nasal bone length (NB), etc., it is necessary to extract the midsagittal plane, which is a plane that divides the subject into two halves symmetrically, from 3D volume data. In addition to extracting the midsagittal plane in fetal examinations, when measuring the subject's corpus callosum, the long-axis length and long-axis cross-sectional area of ​​the left ventricle at end-diastole, etc. on 2D images, it is also necessary to extract planes (hereinafter referred to as standard planes) required for such measurements (hereinafter referred to as standard planes) from 3D volume data.

[0003] Conventionally, standard planes such as the midsagittal plane have been extracted from 3D volume data by users searching for the standard planes in the 3D volume data. In recent years, to reduce the effort required for users to search for standard planes in 3D volume data, standard planes have been automatically extracted from 3D volume data using a technology that searches for standard planes from cross sections in all directions within the 3D volume data or a technology that searches for standard planes using AI (Artificial Intelligence).

[0004] However, when extracting standard planes using a technology that searches all standard planes from cross sections in all directions within 3D volume data, the processing time can be extremely long. Furthermore, when extracting standard planes using a technology that searches for standard planes using AI (Artificial Intelligence), it is difficult to extract standard planes from 3D volume data if there is little difference in the characteristics of the cross section to be extracted and its neighboring cross sections in the 3D volume data. Therefore, it is desirable for users to be able to easily extract standard cross sections of a subject from 3D volume data. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-140588 [Patent Document 2] Japanese Patent Application Laid-Open No. 2016-036594 [Non-patent literature]

[0006] [Non-Patent Document 1] Li Y et al., Standard Plane Detection in 3D Fetal Ultrasound Using an Iterative Transformation Network. Medical Image Computing and Computer Assisted Intervention, MICCAI 2018. Summary of the Invention [Problem to be solved by the invention]

[0007] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to easily extract standard cross sections of a subject from 3D volume data. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be positioned as other problems. [Means for solving the problem]

[0008] A medical processing device according to an embodiment includes an acquisition unit that acquires three-dimensional volume data of a subject, a first cross-section extraction unit that extracts a first cross-section that intersects a reference line of the subject included in the three-dimensional volume data, a second cross-section extraction unit that extracts a second cross-section that intersects the first cross-section from the three-dimensional volume data based on the posture of the subject included in the first cross-section, and an information output unit that outputs information based on the second cross-section. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a block diagram showing an example of the configuration of a medical processing system according to a first embodiment. [Figure 2] 4 is a flowchart for explaining information output processing executed in the medical processing apparatus according to the first embodiment. [Figure 3] FIG. 2 is a diagram showing an example of three-dimensional volume data according to the first embodiment. [Figure 4] FIG. 2 is a diagram illustrating an example of a measurement target of a subject according to the first embodiment. [Figure 5] FIG. 10 is a diagram for explaining a process of extracting a first cross section from a plurality of cross sections. [Figure 6] 1A is a diagram showing one cross section intersecting the reference line, and FIG. 1B is a diagram showing another cross section intersecting the reference line. [Figure 7] 6(a) is a diagram showing the area representing the fetus included in the cross section shown in Fig. 6(a), and (b) is a diagram showing the area representing the fetus included in the cross section shown in Fig. 6(b). [Figure 8]FIG. 10 is a diagram for explaining a process for detecting the posture of a fetus. [Figure 9] FIG. 10 is a diagram for explaining a process of extracting a second cross section. [Figure 10] FIG. 10 is a diagram showing an example of information based on a second cross section. [Figure 11] 10 is a flowchart illustrating an information output process according to Modification 1. [Figure 12] (a) shows the third ventricle, and (b) shows the sacrococcygeal region. [Figure 13] 12(a) is a diagram showing characteristic points of the third ventricle shown in Fig. 12(a), and (b) is a diagram showing characteristic points of the sacrococcygeal region shown in Fig. 12(b). [Figure 14] 10 is a flowchart for explaining information output processing executed in a medical processing apparatus according to a second embodiment. [Figure 15] FIG. 10 is a diagram showing an example of three-dimensional volume data according to the second embodiment. [Figure 16] FIG. 10 is a diagram illustrating an example of a measurement target of a subject according to the second embodiment. [Figure 17] FIG. 1 is a diagram showing an example of a cross section including an eye socket. [Figure 18] FIG. 10 is a diagram illustrating a process for detecting the posture of the head. [Figure 19] FIG. 10 is a diagram for explaining a process of extracting a second cross section. [Figure 20] FIG. 10 is a block diagram showing an example of the configuration of a medical processing system according to Modification 2. [Figure 21] 10 is a flowchart for explaining an information output process executed in a medical processing device according to Modification 2. [Figure 22] FIG. 10 is a diagram showing an example of a bilaterally symmetrical anatomical structure according to Modification 2. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments of a medical processing device and a medical processing method will be described with reference to the drawings. In the following description, components having substantially the same functions and configurations are designated by the same reference numerals, and redundant explanations will be given only when necessary.

[0011] [First embodiment] Fig. 1 is a block diagram showing an example of the configuration of a medical processing system according to the first embodiment. As shown in Fig. 1, the medical processing system 1 is configured to include a medical image diagnostic device 10, a medical image storage device 20, and a medical processing device 30. The medical image diagnostic device 10, the medical image storage device 20, and the medical processing device 30 are connected to each other so as to be able to communicate with each other via an intra-hospital network NW that is a dedicated line within the hospital.

[0012] The medical image diagnostic device 10, the medical image storage device 20, and the medical processing device 30 may be communicably connected to each other via a network using a public line such as the Internet. Although the medical processing device 30 is configured separately from the medical image diagnostic device 10, the medical processing device 30 may be configured integrally with the medical image diagnostic device 10.

[0013] The medical image diagnostic apparatus 10 captures an image of a subject and generates medical data. Then, the medical image diagnostic apparatus 10 transmits the generated medical data to a medical image storage apparatus 20 or a medical processing apparatus 30 via an intra-hospital network NW. For example, the medical image diagnostic apparatus 10 is an ultrasound diagnostic apparatus, an X-ray CT (Computed Tomography) apparatus, an MRI (Magnetic Resonance Imaging) apparatus, an X-ray diagnostic apparatus, a PET apparatus, a SPECT apparatus, or the like. Furthermore, for example, the medical data generated by the medical image diagnostic apparatus 10 is three-dimensional volume data. In the following description, a case where the medical image diagnostic apparatus 10 is an ultrasound diagnostic apparatus will be described as an example.

[0014] The medical image storage device 20 stores medical data generated by the medical image diagnostic device 10 and various data generated by the medical processing device 30. The medical image storage device 20 also transmits the stored medical data and various data to the medical image diagnostic device 10 and to the medical processing device 30 via an intra-hospital network NW. For example, the medical image storage device 20 is an image server such as a PACS (Picture Archiving and Communication System). The medical image storage device 20 may also be realized by a group of servers (cloud) connected to the medical processing system 1 via a network.

[0015] The medical processing device 30 performs various types of information processing related to the subject. Specifically, the medical processing device 30 executes various types of processing using three-dimensional volume data, which is medical data generated by an ultrasound diagnostic device, which is the medical image diagnostic device 10, and three-dimensional volume data transmitted from the medical image storage device 20. For example, the medical processing device 30 is realized by computer equipment such as a server or a workstation.

[0016] As shown in FIG. 1, the medical processing device 30 includes a memory circuit 31, a display 32, an input interface 33, a communication interface , and a processing circuit .

[0017] The memory circuitry 31 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory), a flash memory, a hard disk, an optical disk, etc. The memory circuitry 31 stores, for example, the three-dimensional volume data generated by the medical image diagnostic apparatus 10, the three-dimensional volume data transmitted from the medical image storage apparatus 20, various data generated by the medical processing apparatus 30, etc.

[0018] The display 32 displays various data and information. For example, the display 32 displays a GUI (Graphical User Interface) for accepting various operations from the user. The display 32 is configured by, for example, a liquid crystal display, a CRT (Cathode Ray Tube) display, or the like.

[0019] The input interface 33 accepts various input operations from the user, converts the accepted input operations into electrical signals, and outputs the electrical signals to the processing circuit 35. The input interface 33 may be implemented by, for example, a mouse, keyboard, trackball, manual switch, foot switch, button, joystick, a touchpad that performs input operations by touching the operation surface, a touchscreen that integrates a display screen and a touchpad, a non-contact input interface using an optical sensor, a voice input interface, etc. Note that in this specification, the input interface 33 is not limited to those that have physical operation components such as a mouse and keyboard. For example, an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the device and outputs the electrical signal to a control circuit is also included as an example of the input interface 33.

[0020] The communication interface 34 implements various information communication protocols according to the configuration of the hospital network NW. The communication interface 34 realizes communication with other devices via the hospital network NW in accordance with these various protocols. The medical processing device 30 is connected to the hospital network NW via the communication interface 34, and realizes communication with the medical image diagnostic device 10 and the medical image storage device 20.

[0021] The processing circuitry 35 is an arithmetic circuit that performs various calculations and is configured with a processor such as a CPU or a GPU. The processing circuitry 35, for example, acquires three-dimensional volume data from the medical image diagnostic device 10 or the medical image storage device 20, extracts a first cross section from the three-dimensional volume data, extracts a second cross section from the three-dimensional volume data, and outputs information based on the second cross section. This information based on the second cross section is, for example, an image based on the second cross section extracted from the three-dimensional volume data.

[0022] For this reason, the processing circuitry 35 has an acquisition function 351, a first cross-section extraction function 352, a second cross-section extraction function 353, and an information output function 354. The acquisition function 351 corresponds to the acquisition unit according to this embodiment, the first cross-section extraction function 352 corresponds to the first cross-section extraction unit according to this embodiment, the second cross-section extraction function 353 corresponds to the second cross-section extraction unit according to this embodiment, and the information output function 354 corresponds to the information output unit according to this embodiment.

[0023] Each processing function performed by the acquisition function 351, the first cross-section extraction function 352, the second cross-section extraction function 353, and the information output function 354 is stored in the storage circuitry 31 in the form of a computer-executable program. The processing circuitry 35 is a processor that reads the program from the storage circuitry 31 and executes it to realize the function corresponding to each program. In other words, the processing circuitry 35 in a state in which each program has been read has each function shown in the processing circuitry 35 of FIG. 1. Note that, although FIG. 1 illustrates the acquisition function 351, the first cross-section extraction function 352, the second cross-section extraction function 353, and the information output function 354 being realized by a single processing circuit 35, the processing circuitry 35 may be configured by combining multiple independent processors, and each processor may execute a program to realize these functions.

[0024] The acquisition function 351 acquires three-dimensional volume data of a subject. The first cross-section extraction function 352 extracts a first cross-section that intersects with a reference line of the subject included in the three-dimensional volume data.

[0025] The second cross-section extraction function 353 extracts a second cross-section that intersects with the first cross-section from the three-dimensional volume data based on the posture of the subject included in the first cross-section. The information output function 354 outputs information based on the second cross-section.

[0026] 2 is a flowchart for explaining information output processing executed in the medical processing device 30 according to the first embodiment. In this information output processing, three-dimensional volume data of a fetus, which is a subject, is acquired from the medical image diagnostic device 10 or the medical image storage device 20, a cross section including a predetermined part of the fetus is extracted as a first cross section, a midsagittal cross section is extracted as a second cross section, and information based on the midsagittal cross section is output. For example, the information output processing is processing that is executed when three-dimensional volume data is acquired from the medical image diagnostic device 10 or the medical image storage device 20.

[0027] 2, first, the acquisition function 351 in the processing circuitry 35 of the medical processing device 30 acquires three-dimensional volume data of the fetus (step S11). Specifically, the acquisition function 351 acquires the three-dimensional volume data of the fetus from the medical image diagnostic device 10 or the medical image storage device 20 via the communication interface 34. The acquisition function 351 may directly acquire this three-dimensional volume data of the fetus via the communication interface 34, or the three-dimensional volume data acquired via the communication interface 34 may be temporarily stored in the memory circuitry 31, and then the acquisition function 351 in the processing circuitry 35 may acquire the data from the memory circuitry 31.

[0028] Fig. 3 is a diagram showing an example of three-dimensional volume data according to the first embodiment. As shown in Fig. 3, an acquisition function 351 acquires three-dimensional volume data VD1 of a fetus. In the example shown in Fig. 3, the three-dimensional volume data VD1 of a fetus shows the whole body 50 of the fetus, etc.

[0029] 2, the first plane extraction function 352 in the processing circuitry 35 of the medical processing device 30 extracts a plane including a predetermined region of the fetus (step S13). Specifically, the first plane extraction function 352 extracts a first plane that intersects the reference line and includes the predetermined region of the fetus. More specifically, the first plane extraction function 352 acquires a reference line based on a measurement target of the fetus, which is the subject, included in the three-dimensional volume data VD1, and extracts a plane that intersects the reference line and includes the predetermined region of the fetus as the first plane.

[0030] A method for extracting the first cross section in the medical processing device 30 will be described below.

[0031] In step S13, first, the first plane extraction function 352 detects the fetal measurement target included in the three-dimensional volume data VD1 acquired in step S11. Specifically, the first plane extraction function 352 detects the fetal crown-rump length included in the three-dimensional volume data VD1 as the measurement target. More specifically, the first plane extraction function 352 inputs the three-dimensional volume data VD1 to a trained model, detects the region from the head to the torso in the three-dimensional volume data VD1 using the trained model, and finds the endpoints of the head-rump length from that region using logic, thereby detecting the fetal crown-rump length.

[0032] In other words, the trained model is a model trained to detect a predetermined region in the three-dimensional volume data VD1 when the three-dimensional volume data VD1 is input. In this embodiment, this trained model is a model trained to detect the region from the head to the torso in the three-dimensional volume data VD1 as the predetermined region. The logic is also logic for obtaining information related to the measurement target of the subject from the predetermined region. In this embodiment, this logic is logic for obtaining the end points of the head-rump length from the region from the head to the torso, which is the predetermined region, as information related to the measurement target of the subject.

[0033] Fig. 4 is a diagram showing an example of a measurement target of a subject according to the first embodiment. As shown in Fig. 4, the first cross-section extraction function 352 uses a trained model to detect an area RE1 from the head to the body of the fetus from the three-dimensional volume data VD1, and uses logic to determine two positions that are located on the boundary of the detected area RE1 from the head to the body of the fetus and that are the greatest distance from each other as the endpoints of the crown-rump length, and detects the line segment connecting the two positions as the fetal crown-rump length L1.

[0034] Next, in step S13, the first plane extraction function 352 acquires a reference line based on the measurement object of the fetus included in the three-dimensional volume data VD1. Specifically, the first plane extraction function 352 acquires a reference line based on the crown-rump length L1, which is the measurement object detected from the three-dimensional volume data VD1. More specifically, the first plane extraction function 352 acquires a reference line by setting the crown-rump length L1 in the three-dimensional volume data VD1 based on the measurement object of the crown-rump length L1 detected from the three-dimensional volume data VD1.

[0035] Next, in step S13, the first plane extraction function 352 extracts a first plane that intersects the reference line. Specifically, the first plane extraction function 352 extracts, as the first plane, a plane that includes a predetermined part of the fetus from among a plurality of planes that intersect the reference line acquired based on the crown-rump length L1 at different positions. More specifically, the first plane extraction function 352 extracts, as the first plane, a plane that includes the head, which is a predetermined part of the fetus, from among a plurality of planes that intersect the reference line acquired based on the crown-rump length L1 at different positions. To extract this plane that includes the fetal head, the first plane extraction function 352 extracts, as the first plane, a plane that has the largest area representing the fetus from among the plurality of planes that intersect the reference line at different positions. Furthermore, when extracting multiple cross sections from the three-dimensional volume data VD1, each cross section intersecting the reference line at multiple different positions, the first cross section extraction function 352 extracts each of the multiple cross sections intersecting the reference line at multiple different positions so that the directions in which the cross sections intersect the reference line are the same, i.e., so that the cross sections are parallel to each other.

[0036] FIG. 5 is a diagram illustrating a process for extracting a first plane from multiple planes. As shown in FIG. 5, the first plane extraction function 352 divides the reference line BL1 into multiple equal parts and extracts from the three-dimensional volume data VD1 planes orthogonal to the reference line BL1 at positions P1 estimated to be the fetal head on the entire body 50 of the fetus, among the multiple positions P1 obtained by dividing the reference line BL1 into multiple equal parts. In the example shown in FIG. 5, the first plane extraction function 352 divides the reference line BL1 into four equal parts and extracts from the three-dimensional volume data VD1 planes orthogonal to the reference line BL1 at two different positions P12 and P14 estimated to be the fetal head, among the five positions P1 estimated to be the fetal head. That is, in the example shown in FIG. 5, the first plane extraction function 352 extracts from the three-dimensional volume data VD1 two planes orthogonal to the two different positions P12 and P14 of the reference line BL1, respectively. The two cross sections perpendicular to the two different positions of the reference line BL1 are parallel to each other in the same direction perpendicular to the reference line BL1.

[0037] The first cross-section extraction function 352 extracts from the three-dimensional volume data VD1 cross sections perpendicular to the reference line BL1 at two different positions P12 and P14, among the positions P1 obtained by dividing the reference line BL1 into four equal parts, where the fetal head is estimated to be located. However, the first cross-section extraction function 352 may also extract from the three-dimensional volume data VD1 cross sections perpendicular to the reference line BL1 at each of a plurality of positions P1 obtained by dividing the reference line BL1 into a plurality of equal parts.

[0038] Furthermore, the cross sections intersecting the reference line BL1 at different positions are not limited to those intersecting the reference line BL1 in the same direction. That is, the cross sections intersecting the reference line BL1 at different positions may be those intersecting the reference line BL1 in approximately the same direction, i.e., they may be approximately parallel to each other.

[0039] FIG. 6 is a diagram showing cross sections intersecting a reference line. FIG. 7 is a diagram showing regions representing a fetus included in the cross sections. In the example shown in FIG. 6(a), a region RE11 representing a fetus is depicted in the cross section CS1 at position P12 shown in FIG. 5. In the example shown in FIG. 6(b), a region RE12 representing a fetus is depicted in the cross section CS2 at position P14 shown in FIG. 5. Then, as shown in FIG. 7, the first cross section extraction function 352 uses an ellipse-shaped template to ellipse-fit the regions RE11 and RE12 representing the fetus in each of the multiple cross sections CS1 and CS2 extracted from the three-dimensional volume data VD1. Then, as shown in FIG. 7, the first cross section extraction function 352 extracts, as the first cross section, the cross section CS1, which is the cross section having the largest region representing the fetus among the ellipse-fitted regions RE111 and RE121 representing the fetus in each of the cross sections CS1 and CS2. In this way, the first plane extraction function 352 extracts the plane CS1 that intersects with the reference line BL1 and includes the head of the fetus as the first plane.

[0040] 2, the second plane extraction function 353 in the processing circuitry 35 of the medical processing device 30 extracts a midsagittal plane (step S15). Specifically, the second plane extraction function 353 extracts a midsagittal plane that intersects with the plane CS1 including the fetal head from the three-dimensional volume data VD1 based on the posture of the fetus included in the plane including the fetal head. More specifically, the second plane extraction function 353 detects the orientation of the fetal head as the posture of the fetus, and extracts, as the second plane, a midsagittal plane of the fetus that intersects with the plane including the fetal head (the first plane) and includes the reference line BL1 based on the orientation of the fetal head.

[0041] A method for extracting the second cross section in the medical processing device 30 will be described below.

[0042] In step S15, the second plane extraction function 353 first detects the posture of the fetus included in the plane including the fetal head. Specifically, the second plane extraction function 353 detects the orientation of the fetal head as the posture of the fetus by determining the yaw angle of the fetal head. This yaw angle indicates the tilt (rotation) angle around an axis perpendicular to the cross-section, i.e., an axis parallel to the Z-axis shown in FIG. 3.

[0043] Here, the fetal posture refers to the orientation of the entire fetus, including the orientations of the fetal head and the fetal trunk. However, the second plane extraction function 353 detects the orientation of the fetal head as the orientation of the entire fetus by determining the yaw angle of the fetal head. This is because, for a stationary fetus in the early weeks of pregnancy (for example, up to about 13 weeks of pregnancy), it can be assumed that the yaw angle of the fetal head and the yaw angle of the trunk match. Therefore, by determining the yaw angle of the fetal head, the orientation of the fetal head can be assumed to be the orientation of the entire fetus. Furthermore, since the fetus being the subject is a fetus in the early weeks of pregnancy, the second plane extraction function 353 can detect the orientation of the fetal head as the orientation of the entire fetus.

[0044] 8 is a diagram illustrating the process of detecting the posture of the fetus. As shown in Fig. 8, in order to identify the orientation of the entire fetus from the fetal head, which is a predetermined part, the second plane extraction function 353 determines the major axis LA1 of the ellipse passing through the center C1 of the ellipse in an area RE111 representing the ellipse-fitted fetus in a plane CS1 including the fetal head, and determines the inclination of the major axis LA1 of the ellipse with respect to the horizontal direction of the plane CS1 as the yaw angle of the fetal head, thereby detecting the orientation of the fetal head.

[0045] Next, in step S15, the second plane extraction function 353 extracts a second plane based on the fetal posture included in the plane CS1 including the fetal head. Specifically, the second plane extraction function 353 extracts the second plane based on the orientation of the whole body of the fetus identified from a predetermined region. More specifically, the second plane extraction function 353 extracts a midsagittal plane based on the orientation of the fetal head, which is assumed to be the orientation of the whole body of the fetus.

[0046] 9 is a diagram for explaining the process of extracting the second plane. As shown in Fig. 9, the second plane extraction function 353 rotates the plane by the same angle as the yaw angle of the fetal head (i.e., parallel to the major axis LA1 of the ellipse), obtains a plane PL1 that intersects with the plane CS1 including the fetal head and includes the reference line BL1, and cuts the three-dimensional volume data VD1 with the plane PL1, thereby extracting a fetal midsagittal plane including the reference line BL1 as the second plane.

[0047] 2, the information output function 354 in the processing circuitry 35 of the medical processing device 30 outputs information based on the midsagittal plane (step S17). Specifically, the information output function 354 outputs an image based on the midsagittal plane extracted from the three-dimensional volume data VD1 as the information based on the midsagittal plane.

[0048] Fig. 10 is a diagram showing an example of information based on the second cross section. As shown in Fig. 10, the information output function 354 outputs an image IM1 based on a midsagittal cross section extracted from the three-dimensional volume data VD1 via the display 32. In the example shown in Fig. 10, the image IM1 based on the midsagittal cross section shows the entire body 50 of the fetus. In addition, in the example shown in Fig. 10, the information output function 354 outputs a reference line BL1 acquired based on the fetal crown-rump length L1, superimposed on the image IM1 based on the midsagittal cross section.

[0049] In step S17 described above, the information output function 354 outputs the information based on the midsagittal plane to the display 32, but the output destination of the information based on the midsagittal plane is not limited to the display 32. In other words, the output destination of the information based on the midsagittal plane is arbitrary, and the information output function 354 may output the information based on the midsagittal plane to the medical image diagnostic device 10 via the communication interface 34, for example, or may output the information based on the midsagittal plane to the medical image storage device 20.

[0050] In step S17, the information based on the midsagittal cross section is output, and the information output process ends.

[0051] As described above, the medical processing device 30 detects the fetal head-rump length L1 contained in the fetal three-dimensional volume data VD1, obtains the reference line BL1 based on the detected fetal head-rump length L1, extracts the cross section CS1 containing the fetal head as the first cross section from among multiple cross sections that intersect the fetal reference line BL1 at multiple different positions, extracts a midsagittal cross section that intersects with the cross section CS1 containing the fetal head from the three-dimensional volume data VD1 based on the orientation of the fetus contained in the extracted cross section CS1 containing the fetal head, and outputs an image based on the midsagittal cross section to the display 32. This allows the user to easily generate the fetal midsagittal cross section, which is one of the standard cross sections.

[0052] In step S13 of the information output processing described above, the first plane extraction function 352 uses an elliptical template to ellipse-fit the regions RE11 and RE12 representing the fetus in each of the multiple planes CS1 and CS2 extracted from the three-dimensional volume data VD1, and extracts the plane CS1, which is the plane with the largest region representing the fetus among the ellipse-fitted regions RE111 and RE121 representing the fetus in each of the planes CS1 and CS2, as the first plane. However, in step S13 of the information output processing described above, the first plane extraction function 352 may execute a segmentation process to extract the regions RE11 and RE12 representing the fetus in each of the multiple planes CS1 and CS2 extracted from the three-dimensional volume data VD1, and extract the plane CS1, which is the plane with the largest region representing the fetus among the regions RE111 and RE121 representing the fetus in each of the extracted planes CS1 and CS2, as the first plane.

[0053] In this case, in step S15 of the information output process described above, the second cross-section extraction function 353 may use an ellipse-shaped template to ellipse-fit the region RE111 representing the fetus in the cross-section CS1 including the fetal head to the shape of the head, determine the major axis LA1 of the ellipse passing through the center C1 of the ellipse in the ellipse-fitted region RE111 representing the fetus in the cross-section CS1 including the fetal head, and determine the inclination of the major axis LA1 of the ellipse with respect to the horizontal direction of the cross-section CS1 as the yaw angle of the fetal head, thereby detecting the orientation of the fetal head.

[0054] Furthermore, although the first plane extraction function 352 executes a segmentation process to extract regions RE11 and RE12 representing the fetus in each of the multiple planes CS1 and CS2 extracted from the three-dimensional volume data VD1, the method for extracting regions RE11 and RE12 representing the fetus in each of the multiple planes CS1 and CS2 is not limited to this. That is, any method can be used to extract regions RE11 and RE12 representing the fetus. For example, the first plane extraction function 352 may extract regions RE11 and RE12 representing the fetus using an algorithm that extracts regions RE11 and RE12 representing the fetus, or may extract regions RE11 and RE12 representing the fetus using a trained model that inputs planes extracted from the three-dimensional volume data VD1 and extracts regions RE11 and RE12 representing the fetus included in the planes.

[0055] Furthermore, although the predetermined part of the fetus is the head of the fetus, the predetermined part of the fetus is not limited to the head of the fetus. In other words, the predetermined part of the fetus is arbitrary, and the predetermined part of the fetus may be a part other than the head of the fetus.

[0056] Furthermore, although the fetus being the subject is assumed to be a fetus in the first week of pregnancy, the subject fetus is not limited to a fetus in the first week of pregnancy. In other words, the subject fetus may be any fetus, and may be a fetus in a week after the first week of pregnancy. In this case, in step S15 of the information output process, the second plane extraction function 353 may detect the posture of the fetus, for example, by determining the yaw angle of the fetus's head and torso.

[0057] [Variation 1] In the above-described medical processing device 30, the first plane extraction function 352 detects the fetal crown-rump length L1 as the reference line of the subject contained in the three-dimensional volume data VD1, and extracts the first plane that intersects with the reference line BL1 obtained based on the detected fetal crown-rump length L1. However, it is also possible to obtain a line connecting feature points of multiple anatomical structures of the fetus contained in the three-dimensional volume data VD1 as the reference line of the subject contained in the three-dimensional volume data VD1, and extract the first plane that intersects with the line connecting feature points of multiple anatomical structures of the fetus. Note that the configurations of the medical processing system 1 and the medical processing device 30 are the same as those in Figure 1, so their explanation will be omitted.

[0058] FIG. 11 is a flowchart illustrating information output processing according to Modification 1, and corresponds to FIG. 2. In the information output processing according to Modification 1, three-dimensional volume data of a fetus, which is a subject, is also acquired from the medical image diagnostic device 10 or the medical image storage device 20, a cross section including a predetermined part of the fetus is extracted as a first cross section, a midsagittal cross section is extracted as a second cross section, and information based on the midsagittal cross section is output. For example, the information output processing is processing that is executed when three-dimensional volume data is acquired from the medical image diagnostic device 10 or the medical image storage device 20. Note that the processing of step S11 shown in FIG. 11 is the same as that in FIG. 2, and therefore description thereof will be omitted.

[0059] 11, the first plane extraction function 352 in the processing circuitry 35 of the medical processing device 30 extracts a plane including a predetermined region of the fetus (step S13a). Specifically, the first plane extraction function 352 according to this modification detects a plurality of anatomical structures of the fetus, which is the subject, included in the three-dimensional volume data VD1, acquires lines connecting feature points of the plurality of anatomical structures as reference lines of the subject included in the three-dimensional volume data VD1, and extracts, as the first plane, a plane including a predetermined region of the fetus from among a plurality of planes that intersect at a plurality of different positions on the line connecting the plurality of anatomical structures acquired as the reference line.

[0060] A method for extracting the first cross section in the medical processing device 30 according to this modification will be described below.

[0061] In step S13a, first, the first plane extraction function 352 according to the first modification detects a plurality of anatomical structures of the fetus included in the three-dimensional volume data VD1 acquired in step S11. Specifically, the first plane extraction function 352 according to this modification detects a plurality of anatomical structures of the fetus included in the three-dimensional volume data VD1 by executing a segmentation process.

[0062] Although the first plane extraction function 352 according to this modification detects multiple anatomical structures of a fetus included in the three-dimensional volume data VD1 by executing a segmentation process, the method for detecting multiple anatomical structures of a fetus included in the three-dimensional volume data VD1 is not limited to this. That is, any method can be used to detect multiple anatomical structures of a fetus included in the three-dimensional volume data VD1. For example, the first plane extraction function 352 may detect multiple anatomical structures using an algorithm for detecting anatomical structures, or may detect multiple anatomical structures using a trained model that receives the three-dimensional volume data VD1 as input and detects multiple anatomical structures included in the three-dimensional volume data VD1.

[0063] 12A and 12B are diagrams illustrating anatomical structures. As shown in Fig. 12A, the first plane extraction function 352 according to this modification detects the third ventricle TV of the fetus included in the three-dimensional volume data VD1 as one of the anatomical structures. As shown in Fig. 12B, the first plane extraction function 352 according to this modification detects the sacrococcygeal region SC as one of the anatomical structures.

[0064] Next, in step S13a, the first cross-section extraction function 352 according to the first modification identifies feature points of the plurality of anatomical structures. Specifically, the first cross-section extraction function 352 according to the first modification identifies the centers of the plurality of anatomical structures as feature points of the plurality of anatomical structures.

[0065] Note that the first plane extraction function 352 according to Modification 1 identifies the center of each of the plurality of anatomical structures as a feature point of the plurality of anatomical structures, but the position of each of the plurality of anatomical structures identified by the first plane extraction function 352 according to Modification 1 is not limited to the center. In other words, the position of each of the plurality of anatomical structures identified by the first plane extraction function 352 according to Modification 1 is arbitrary, and the first plane extraction function 352 according to Modification 1 may identify a position other than the center of each of the plurality of anatomical structures.

[0066] 13A and 13B are diagrams illustrating feature points of anatomical structures. As shown in Fig. 13A, the first cross-section extraction function 352 according to the first modification identifies the center TV_C of the third ventricle TV, which is one of the detected anatomical structures, as a feature point. As shown in Fig. 13B, the first cross-section extraction function 352 according to the first modification identifies the center SC_C of the sacrococcygeal region SC, which is one of the detected anatomical structures, as a feature point.

[0067] Next, in step S13a, the first cross-section extraction function 352 according to the first modification acquires a line connecting feature points of a plurality of anatomical structures as a reference line. Specifically, in the example shown in Fig. 13, the first cross-section extraction function 352 according to the first modification acquires a line connecting the center TV_C of the third ventricle TV and the center SC_C of the sacrococcygeal region SC as a reference line.

[0068] The process of step S13a after obtaining the line connecting the feature points of the plurality of anatomical structures as the reference line is the same as the process of step S13, and therefore a description thereof will be omitted. Furthermore, the process of steps S15 and S17 after step S13a is the same as the process of the first embodiment, and therefore a description thereof will be omitted. Then, in step S17, information based on the midsagittal section is output, and the information output process according to this modification 1 is completed.

[0069] As described above, in the medical processing device 30 according to the first modification, a line connecting feature points of a plurality of anatomical structures included in the three-dimensional volume data VD1 is acquired as a reference line, a cross section including the head of the fetus is extracted as a first cross section from among a plurality of cross sections that intersect with the line connecting feature points of the acquired plurality of anatomical structures at different positions, a midsagittal cross section that intersects with the first cross section is extracted from the three-dimensional volume data VD1 based on the orientation of the fetus included in the extracted first cross section, and an image based on the midsagittal cross section is output to the display, so that the user can easily generate a midsagittal cross section of the fetus, which is one of the standard cross sections.

[0070] Second Embodiment In the medical processing device 30 according to the first embodiment described above, when the subject is a fetus and the measurement target is the crown-rump length, an image based on a midsagittal plane of the fetus is output as information based on the second plane. However, the subject, measurement target, and information based on the second plane are not limited to this. In the second embodiment, a medical processing device 30 is described that outputs an image based on a horizontal plane of the head as information based on the second plane when the subject is the head and the measurement target is the corpus callosum. Note that the configurations of the medical processing system 1 and the medical processing device 30 according to the second embodiment are the same as those shown in FIG. 1 according to the first embodiment described above, and therefore description thereof will be omitted.

[0071] 14 is a flowchart illustrating information output processing executed in the medical processing device 30 according to the second embodiment, and corresponds to FIG. 2. In this information output processing, three-dimensional volume data of the subject's head is acquired from the medical image diagnostic device 10 or the medical image storage device 20, a cross section including the eye socket is extracted as a first cross section, a horizontal cross section of the head is extracted, and information based on the horizontal cross section of the head is output. For example, the information output processing is processing that is executed when three-dimensional volume data of the head is acquired from the medical image diagnostic device 10 or the medical image storage device 20.

[0072] 14, first, the acquisition function 351 in the processing circuitry 35 of the medical processing device 30 acquires three-dimensional volume data of the head (step S21). Specifically, the acquisition function 351 acquires three-dimensional volume data of the head, which is the subject, from the medical image diagnostic apparatus 10 or the medical image storage device 20 via the communication interface 34. This three-dimensional volume data of the head may be acquired directly by the acquisition function 351 via the communication interface 34, or the three-dimensional volume data acquired via the communication interface 34 may be temporarily stored in the memory circuitry 31, and then the acquisition function 351 in the processing circuitry 35 may acquire the data from the memory circuitry 31.

[0073] 15 is a diagram showing an example of three-dimensional volume data according to the second embodiment. As shown in FIG.

[0074] 14, the first plane extraction function 352 in the processing circuitry 35 of the medical processing device 30 extracts a plane including the eye socket (step S23). Specifically, the first plane extraction function 352 extracts a plane that intersects the reference line and includes the eye socket as the first plane. More specifically, the first plane extraction function 352 acquires a reference line based on the measurement target of the head 60, which is the subject, included in the three-dimensional volume data VD1a, and extracts a plane that intersects the reference line and includes the eye socket as the first plane.

[0075] A method for extracting the first cross section in the medical processing device 30 will be described below.

[0076] In step S23, first, the first cross-section extraction function 352 detects the measurement target of the head included in the three-dimensional volume data VD1a acquired in step S21. Specifically, the first cross-section extraction function 352 detects the corpus callosum of the head 60 included in the three-dimensional volume data VD1a as the measurement target. More specifically, the first cross-section extraction function 352 inputs the three-dimensional volume data VD1a into a trained model, and the trained model outputs the corpus callosum of the head 60, thereby detecting the corpus callosum of the head 60. In other words, the trained model is a model that has been trained to output the corpus callosum of the head 60 in the three-dimensional volume data VD1a in response to the input of the three-dimensional volume data VD1a.

[0077] Fig. 16 is a diagram showing an example of a measurement target of a subject according to the second embodiment. As shown in Fig. 16, the first cross-section extraction function 352 detects the corpus callosum CC of the head 60 from the three-dimensional volume data VD1a using a trained model.

[0078] Next, in step S23, the first cross-section extraction function 352 acquires a reference line based on the measurement target of the head included in the three-dimensional volume data VD1a. Specifically, the first cross-section extraction function 352 acquires the reference line based on the corpus callosum CC, which is the measurement target detected from the three-dimensional volume data VD1a. More specifically, the first cross-section extraction function 352 acquires the reference line by linearly approximating the corpus callosum CC based on the corpus callosum CC detected from the three-dimensional volume data VD1a. By acquiring the reference line, the first cross-section extraction function 352 determines the orientation of the corpus callosum CC, i.e., the tilt angle (pitch angle) and yaw angle of the head 60. Here, the tilt angle (pitch angle) refers to the tilt angle around an axis parallel to the Y-axis shown in FIG. 15. The yaw angle refers to the tilt angle around an axis parallel to the Z-axis shown in FIG. 15.

[0079] Next, in step S23, the first plane extraction function 352 extracts a first plane that intersects the reference line. Specifically, the first plane extraction function 352 extracts a plane that intersects the reference line acquired based on the corpus callosum CC and includes the orbit as the first plane. More specifically, the first plane extraction function 352 extracts a plane that includes the orbit as the first plane from among multiple planes that intersect the reference line acquired based on the corpus callosum CC at multiple different positions. Furthermore, when extracting multiple planes that intersect the reference line at multiple different positions from the three-dimensional volume data VD1a, the first plane extraction function 352 extracts the multiple planes that intersect the reference line at multiple different positions so that the directions of intersecting the reference line are the same, i.e., so that the planes are parallel to each other.

[0080] Fig. 17 is a diagram showing an example of a cross section including the orbit. As shown in Fig. 17, the first cross section extraction function 352 extracts a cross section CS3 including the orbit ES as a first cross section from a plurality of cross sections that are orthogonal to a plurality of different positions on the reference line based on the corpus callosum CC.

[0081] 14, the second plane extraction function 353 in the processing circuitry 35 of the medical processing device 30 extracts a head horizontal plane (step S25). Specifically, the second plane extraction function 353 extracts a head horizontal plane that intersects with the plane CS3 including the orbits ES from the three-dimensional volume data VD1a based on the head posture included in the plane CS3 including the orbits ES. More specifically, the second plane extraction function 353 extracts a coronal plane including a pair of left and right orbits from the three-dimensional volume data VD1a based on the plane CS3 including the orbits, and extracts a head horizontal plane that intersects with the coronal plane as a second plane based on the direction connecting the pair of left and right orbits included in the coronal plane.

[0082] A method for extracting the second cross section in the medical processing device 30 will be described below.

[0083] In step S25, first, the second plane extraction function 353 detects the posture of the head included in the plane CS3 including the orbits ES. Specifically, the second plane extraction function 353 extracts a coronal plane including the pair of left and right orbits ES from the three-dimensional volume data VD1a based on the plane CS3 including the orbits ES extracted as the first plane in step S23, and detects the orientation of the head 60 as the posture of the head 60 by finding the direction connecting the pair of left and right orbits ES included in the coronal plane.

[0084] FIG. 18 is a diagram illustrating a process for detecting the posture of the head 60. As shown in FIG. 18, the second plane extraction function 353 extracts a coronal plane COS including the pair of left and right orbits ES from the 3D volume data VD1a based on the plane including the eye orbits ES extracted as the first plane. To identify the orientation of the head 60 from the pair of eye orbits ES, which is a predetermined region, the second plane extraction function 353 determines a direction D1 connecting the pair of eye orbits ES included in the coronal plane COS and determines the tilt of the direction D1 connecting the pair of eye orbits ES included in the coronal plane COS relative to the horizontal direction of the coronal plane COS as the roll angle of the head 60, thereby detecting the orientation of the head 60. This head roll angle refers to the tilt angle around an axis parallel to the X-axis shown in FIG. 15.

[0085] Next, in step S25, the second plane extraction function 353 extracts a second plane based on the orientation of the head 60 included in the plane CS3 including the eye orbits ES. Specifically, the second plane extraction function 353 extracts the second plane based on the orientation of the head identified from a predetermined region. More specifically, the second plane extraction function 353 extracts a horizontal plane of the head as the second plane based on the orientation of the head 60, which is the orientation of the head 60 identified from the pair of eye orbits ES, which are the predetermined regions.

[0086] 19 is a diagram for explaining the process of extracting the second cross section. As shown in Fig. 19, the second cross section extraction function 353 rotates the head 60 by the same angle as the roll angle (i.e., parallel to the direction D1 connecting the pair of left and right orbits ES included in the coronal cross section COS), obtains a plane PL1a including a reference line BL1a that intersects with the coronal cross section COS extracted based on the cross section CS3 including the orbits ES, and extracts a horizontal cross section of the head including the reference line BL1a as the second cross section by cutting the three-dimensional volume data VD1a with the plane PL1a.

[0087] 14, the information output function 354 in the processing circuitry 35 of the medical processing device 30 outputs information based on the head horizontal cross section (step S27). Specifically, the information output function 354 outputs an image based on the head horizontal cross section extracted from the three-dimensional volume data VD1a to the display 32 as the information based on the head horizontal cross section.

[0088] In step S27 described above, the information output function 354 outputs the information based on the horizontal head section to the display 32, but the output destination of the information based on the horizontal head section is not limited to the display 32. In other words, the output destination of the information based on the horizontal head section is arbitrary, and the information output function 354 may output the information based on the horizontal head section to the medical image diagnostic apparatus 10 via the communication interface 34, for example, or may output the information based on the horizontal head section to the medical image storage apparatus 20.

[0089] In step S27, the information based on the horizontal cross section of the head is output, and the information output process ends.

[0090] As described above, the medical processing device 30 detects the corpus callosum CC of the head 60 contained in the three-dimensional head volume data VD1a, extracts the cross section CS3 including the orbit ES as the first cross section from among multiple cross sections that intersect with the reference line BL1a based on the detected corpus callosum CC of the head 60, extracts a horizontal head cross section that intersects with the cross section CS3 including the orbit ES and includes the reference line BL1a from the three-dimensional volume data VD1a based on the posture of the head 60 contained in the extracted cross section CS3 including the orbit ES, and outputs an image based on the horizontal head cross section to the display 32, allowing the user to easily generate a horizontal head cross section, which is one of the standard cross sections.

[0091] [Variation 2] In the medical processing device 30 according to the first and second embodiments described above, it is also possible to change the position of the reference line based on symmetrical anatomical structures included in the first cross section, and extract the first cross section that intersects with the changed reference line. Hereinafter, a case where this modification is applied to the first embodiment will be referred to as Modification 2, and differences from the first embodiment described above will be described. Note that, although a case where this modification is applied to the first embodiment will be described below, this modification can also be applied to Modification 1 and the second embodiment described above.

[0092] Fig. 20 is a block diagram showing an example of the configuration of a medical processing system 1 according to Modification 2, and corresponds to Fig. 1. As shown in Fig. 20, in the medical processing device 30 in the medical processing system 1 according to this modification, the first cross-section extraction function of the processing circuitry 35 is different from that of the first embodiment described above, and is therefore referred to as a first cross-section extraction function 352a. Note that the configuration and functions other than the first cross-section extraction function 352a are the same as those of Fig. 1 of the first embodiment described above, and therefore description thereof will be omitted.

[0093] The first plane extraction function 352a according to this modification changes the position of the reference line BL1 based on a bilaterally symmetrical anatomical structure included in the plane CS1 including the fetal head, and extracts the plane CS1 including the fetal head that intersects with the changed reference line BL1. Here, the bilaterally symmetrical anatomical structure refers to bilaterally symmetrical organs and tissues such as the anterior horn of the lateral ventricle, kidneys, brain, hands, feet, ears, eyes, and lungs. In the following description, the anterior horn of the lateral ventricle is used as an example of the bilaterally symmetrical anatomical structure.

[0094] FIG. 21 is a flowchart illustrating information output processing executed in a medical processing device 30 according to Modification 2, and corresponds to FIG. 2. In the information output processing according to Modification 2, three-dimensional volume data of a fetus, which is a subject, is also acquired from the medical image diagnostic device 10 or the medical image storage device 20, a cross section including a predetermined part of the fetus is extracted as a first cross section, a midsagittal cross section is extracted as a second cross section, and information based on the midsagittal cross section is output. For example, the information output processing is processing executed when three-dimensional volume data is acquired from the medical image diagnostic device 10 or the medical image storage device 20. Note that the processing of step S11 shown in FIG. 21 is the same as that in FIG. 2, and therefore description thereof will be omitted.

[0095] Next, as shown in FIG. 21 , the first plane extraction function 352a in the processing circuitry 35 of the medical processing device 30 extracts a first plane (step S13b). Specifically, the first plane extraction function 352a extracts a first plane that intersects the reference line and includes a predetermined region of the fetus. More specifically, the first plane extraction function 352a acquires a reference line based on the crown-rump length of the subject fetus, which is included in the three-dimensional volume data VD1, and extracts a plane that intersects the reference line and includes the predetermined region of the fetus as the first plane. The first plane extraction function 352a then detects symmetrical anatomical structures included in the plane that includes the predetermined region of the fetus, determines whether to change the position of the reference line BL1, and if the position of the reference line BL1 is to be changed, changes the position of the reference line based on the symmetrical anatomical structures included in the first plane, and extracts a first plane that intersects the changed reference line.

[0096] Hereinafter, a method for extracting the first cross section in the medical processing device 30 according to this modification will be described. Note that the process up to the point where the first cross section extraction function 352a extracts, as the first cross section, a cross section including a predetermined part of the fetus from among a plurality of cross sections that intersect at a plurality of different positions on the reference line acquired based on the crown-rump length, is the same as in the first embodiment described above, and therefore description thereof will be omitted.

[0097] Next, in step S13b, the first plane extraction function 352a detects bilaterally symmetrical anatomical structures included in the extracted plane including the fetal head, and determines whether or not to change the position of the reference line BL1. Specifically, the first plane extraction function 352a according to this modification uses a segmentation process to detect bilaterally symmetrical anatomical structures included in the plane including the fetal head, and determines whether or not to change the position of the reference line by determining whether or not the detected bilaterally symmetrical anatomical structures included in the plane including the fetal head are bilaterally symmetrical in the plane including the fetal head.

[0098] Although the first plane extraction function 352 according to this modification detects multiple anatomical structures of a fetus included in the three-dimensional volume data VD1 by executing a segmentation process, the method for detecting multiple anatomical structures of a fetus included in the three-dimensional volume data VD1 is not limited to this. That is, any method can be used to detect multiple anatomical structures of a fetus included in the three-dimensional volume data VD1. For example, the first plane extraction function 352 may detect multiple anatomical structures using an algorithm for detecting anatomical structures, or may detect multiple anatomical structures using a trained model that receives the three-dimensional volume data VD1 as input and detects multiple anatomical structures included in the three-dimensional volume data VD1.

[0099] 22 is a diagram showing an example of a bilaterally symmetrical anatomical structure according to Modification 2. As shown in Fig. 22, the first plane extraction function 352a according to this modification detects the anterior horns of the lateral ventricles LV included in the plane CS1 including the fetal head as a bilaterally symmetrical anatomical structure. The first plane extraction function 352a then determines whether or not to change the position of the reference line BL1 by determining whether or not the anterior horns of the left and right lateral ventricles LV included in the detected plane CS1 including the fetal head are bilaterally symmetrical on the first plane.

[0100] Although the first plane extraction function 352a according to this modification detects the anterior horn of the lateral ventricle LV as a bilaterally symmetrical anatomical structure from the plane CS1 including the fetal head, the bilaterally symmetrical anatomical structure detected by the first plane extraction function 352a from the plane CS1 including the fetal head is not limited to the anterior horn of the lateral ventricle. In other words, the bilaterally symmetrical anatomical structure detected by the first plane extraction function 352a from the plane CS1 including the fetal head is arbitrary.

[0101] If the anterior horns of the left and right lateral ventricles LV are symmetrical on the cross section including the fetal head, the first cross section extraction function 352a according to this modification does not change the position of the reference line BL1. On the other hand, if the anterior horns of the left and right lateral ventricles LV are not symmetrical on the cross section including the fetal head, the first cross section extraction function 352a according to this modification changes the position of the reference line BL1 based on the anterior horns of the left and right lateral ventricles LV included in the cross section including the fetal head, and extracts a cross section including the fetal head that intersects with the changed reference line BL1.

[0102] The processes of steps S15 and S17 after step S13b are the same as those of the first embodiment, and therefore will not be described again. Then, in step S17, information based on the second cross section is output, and the information output process ends.

[0103] As described above, in the medical processing device 30 of this modified example, the position of the reference line BL1 is changed based on the bilaterally symmetrical anatomical structures contained in the first cross section, and the first cross section that intersects with the changed reference line is extracted, so that a more accurate midsagittal cross section of the fetus can be easily generated.

[0104] [Variation 3] In the medical processing system 1 according to the first and second embodiments, and modified examples 1 and 2 described above, the configuration of the medical processing device 30 can also be applied to the medical image diagnostic device 10 and the medical image storage device 20. In that case, for example, functions equivalent to the above-described acquisition function 351, first cross-section extraction functions 352, 352a, second cross-section extraction function 353, and information output function 354 are implemented in the processing circuit provided in the medical image diagnostic device 10 and the processing circuit provided in the medical image storage device 20.

[0105] [Other Modifications] In the medical processing system 30 of the first and second embodiments and the first to third modifications, in steps S13, S13a, and S13b of the information output process, the first cross-section extraction functions 352 and 352a extract the first cross-section perpendicular to the reference lines BL1 and BL1a. However, the first cross-section does not have to be perpendicular to the reference lines BL1 and BL1a. That is, the angle formed by the reference lines BL1 and BL1a and the first cross-section is arbitrary, and the first cross-section extraction functions 352 and 352a only need to extract the first cross-section that intersects with the reference lines BL1 and BL1a.

[0106] Furthermore, in the medical processing device 30 of the medical processing system 1 according to the first and second embodiments and Modifications 1 to 3 described above, in steps S13, S13a, and S13b of the information output process, the first cross-section extraction functions 352, 352a detect the measurement target of the subject using a trained model or logic, but the method of detecting the measurement target of the subject is not limited to this. That is, any method of detecting the measurement target of the subject may be used. For example, the first cross-section extraction function 352 may detect the measurement target of the subject from the three-dimensional volume data VD1, VD1a using only logic for detecting the measurement target of the subject, or may receive an input operation related to the detection of the measurement target of the subject from a user and detect the measurement target of the subject from the three-dimensional volume data VD1, VD1a in accordance with the received input operation.

[0107] Although the first embodiment and the second embodiment are described above with reference to an example in which the subject is a fetus and a head, respectively, the subject is not limited to this. That is, the subject may be any subject, and may be, for example, a heart.

[0108] In the second embodiment described above, the second plane extraction function 353 rotates the head 60 by the same angle as the roll angle, obtains a plane PL1a that intersects with the coronal plane COS extracted based on the plane CS3 including the orbits ES and includes the reference line BL1a, and cuts the three-dimensional volume data VD1a with the plane PL1a to extract a head horizontal plane including the reference line BL1a as the second plane. However, the head horizontal plane is not limited to including the reference line BL1a. In other words, the second plane extraction function 353 may rotate the head 60 by the same angle as the roll angle, obtains a plane that intersects with the coronal plane COS extracted based on the plane CS3 including the orbits ES and does not include the reference line BL1a, and cuts the three-dimensional volume data VD1a with the plane to extract a head horizontal plane not including the reference line BL1a as the second plane.

[0109] The term "processor" used in the above description refers to a circuit such as a central processing unit (CPU), a graphics processing unit (GPU), 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)). The processor realizes its functions by reading and executing a program stored in the memory circuit 31. Instead of storing the program in the memory circuit 31, the processor may be configured so that the program is directly embedded in its circuit. In this case, the processor realizes its functions by reading and executing the program embedded in the circuit. The processor is not limited to being configured as a single circuit, but may also be configured as a single processor by combining multiple independent circuits to realize its functions. Furthermore, multiple components in FIG. 1 may be integrated into a single processor to realize its functions.

[0110] Although several embodiments have been described above, these embodiments are presented only as examples and are not intended to limit the scope of the invention. The novel apparatus and method described herein may be embodied in various other forms. Furthermore, various omissions, substitutions, and modifications may be made to the forms of the apparatus and method described herein without departing from the spirit of the invention. The appended claims and their equivalents are intended to cover such forms and modifications that fall within the scope and spirit of the invention. [Explanation of symbols]

[0111] 1... medical processing system, 10... medical image diagnostic device, 20... medical image storage device, 30... medical processing device, 31... memory circuit, 32... display, 33... input interface, 34... communication interface, 35... processing circuit, 351... acquisition function, 352, 352a... first cross-section extraction function, 353... second cross-section extraction function, 354... information output function

Claims

1. an acquisition unit for acquiring three-dimensional volume data of a subject; a first cross-section extraction unit that extracts a first cross-section that intersects a reference line of the subject included in the three-dimensional volume data; a second cross-section extraction unit that extracts a second cross-section that intersects with the first cross-section from the three-dimensional volume data, based on the posture of the subject included in the first cross-section; an information output unit that outputs information based on the second cross section, Medical processing equipment.

2. the first cross-section extraction unit extracts a first cross-section that intersects the reference line and includes a predetermined part of the subject; the second cross-section extraction unit extracts the second cross-section based on the posture of the subject identified from the predetermined region. The medical processing device of claim 1 .

3. the first cross-section extraction unit extracts, as the first cross-section, a cross-section including a predetermined part of the subject from among a plurality of cross-sections that intersect the reference line at a plurality of different positions; The medical processing device of claim 2 .

4. the subject is a fetus; the first cross-section extraction unit acquires the reference line based on a measurement target of the fetus included in the three-dimensional volume data, and extracts the first cross section intersecting the reference line; The medical processing device of claim 1 .

5. The measurement target is crown rump length (CRL), the first cross-section extraction unit acquires the reference line based on a crown-rump length detected from the three-dimensional volume data. The medical processing device of claim 4 .

6. the first plane extraction unit extracts a plane that intersects the reference line and includes the head of the fetus as the first plane; the second plane extraction unit extracts the second plane based on the posture of the fetus identified from the head of the fetus. The medical processing device of claim 5 .

7. the first plane extraction unit extracts, as the first plane including the head of the fetus, a plane having the largest area representing the fetus from among a plurality of planes that respectively intersect the reference line at a plurality of different positions; The medical processing device of claim 6 .

8. the second plane extraction unit detects the orientation of the head of the fetus as the posture of the fetus, and extracts, as the second plane, a midsagittal plane of the fetus that intersects with the first plane and includes the reference line based on the orientation of the head of the fetus. The medical processing device of claim 6 .

9. the subject is a head, the first cross-section extraction unit acquires the reference line based on a measurement target of the head included in the three-dimensional volume data. The medical processing device of claim 1 .

10. the measurement target is the corpus callosum, the first cross-section extraction unit acquires the reference line based on the corpus callosum detected from the three-dimensional volume data. The medical processing device of claim 9 .

11. the first cross-section extraction unit extracts a cross-section that intersects with the reference line and includes an eye socket as the first cross-section; The medical processing device of claim 10.

12. the second cross-section extraction unit extracts a coronal cross-section including the pair of left and right orbits from the three-dimensional volume data based on the cross-section including the orbits, and extracts a horizontal head cross-section intersecting the coronal cross-section as the second cross-section based on a direction connecting the pair of left and right orbits included in the coronal cross-section. The medical processing device of claim 11 .

13. the first cross-section extraction unit detects a plurality of anatomical structures of the subject and acquires a line connecting feature points of the plurality of anatomical structures as the reference line; The medical processing device of claim 1 .

14. the first cross-section extraction unit changes a position of the reference line based on a bilaterally symmetrical anatomical structure included in the first cross-section, and extracts the first cross-section intersecting the changed reference line; The medical processing device of claim 1 .

15. the first cross-section extraction unit inputs the three-dimensional volume data into a trained model, detects a predetermined region of the subject using the trained model, and obtains information related to a measurement target of the subject from the predetermined region using logic, thereby detecting the measurement target of the subject. The medical processing device of claim 1 .

16. an acquisition step of acquiring three-dimensional volume data of a subject; a first plane extraction step of extracting a first plane that intersects a reference line of the subject included in the three-dimensional volume data; a second plane extraction step of extracting a second plane intersecting the first plane from the three-dimensional volume data based on the posture of the subject included in the first plane; and an information output step of outputting information based on the second cross section. Medical processing methods.

Images