JP5479138B2 - MEDICAL IMAGE DISPLAY DEVICE, MEDICAL IMAGE DISPLAY METHOD, AND PROGRAM THEREOF - Google Patents

Description

  The present invention relates to a medical image display device that displays a cross-sectional image of a medical image, a medical image display method, and a program thereof.

  Conventionally, in the field of medical images, modalities using various techniques such as an X-ray CT apparatus, an ultrasonic diagnostic apparatus, and an MRI apparatus are used in addition to an X-ray imaging apparatus. When observing medical images obtained by these devices, doctors often want to observe images of predetermined cross sections defined anatomically as shown in medical books and the like. .

  However, since medical images are taken under various conditions depending on the disease, the purpose of the examination, the patient's state at the time of taking an image, etc., the difference in the shooting range and FOV (Field Of View) in the body axis direction for each image, Due to differences in the positional relationship and orientation of the patient's organs in the image due to differences in the patient's body position and individual differences, doctors and others have taken time to search for a predetermined cross-sectional image.

  Therefore, when it is desired to observe the image of the predetermined cross section, doctors and the like first use the axial cross section, the coronal cross section, the sagittal cross section, and the like to check whether the cross section is included in the medical image. The approximate position of the cross section is searched from the image, and then the angle and the position are finely adjusted to obtain the image of the predetermined cross section.

  Patent Document 1 describes an image display device that allows a user to search for an image that the user wants to observe by operating the displayed image using an input device such as a pointing device.

JP 2009-022476 A

  Conventionally, a doctor or the like manually searches for an image of the predetermined cross-section from medical images, which takes time and effort, and places a burden on the doctor or the like. Also, when dealing with images taken over a wide range, or when referring to images of similar cases or past images, it is not clear whether an image of a predetermined cross section is included in an organ other than the main purpose of the examination. There was a case.

  Therefore, the present invention has been made in view of such conventional problems, and a medical image display device capable of automatically specifying a desired cross section included in a medical image and displaying the image of the specified cross section. It is an object of the present invention to provide a medical image display method and a program thereof.

To achieve the above object, the present invention provides a medical image display apparatus, an image acquiring unit for acquiring 3-dimensional image image of the medical image, by analyzing the 3-dimensional image image, one or more pre among a defined reference plane, the cross section specifying unit configured to specify a cross section corresponding to the reference cross-section included in the medical image, the three-dimensional image image or al, the cross section of the prior Kidan surface on which the cross-section identifying unit has identified an image generator for generating the images, from among the cross section specified by the section specifying unit, a selection unit for selecting the cross-section indicated by the user, the image generation unit generates, and the selection and a display control unit for displaying on the display unit sectional image image of the cross-section selected by section prohibits the selection by the selection unit of the cross section the section specifying unit could not be identified, the The reference cross section is the apex left ventricular long axis , The apex four-chamber cross section, the apex two-chamber cross section, and the left ventricular short-axis cross section, wherein the cross-section specifying unit analyzes the three-dimensional image of the medical image to define the reference cross section The apex, the base, the lower part of the aortic valve, and the upper part of the aortic valve are detected, and the plane defined by the apex, the base and the lower part of the aortic valve corresponds to the left ventricular long-axis cross section of the apex The plane specified by the cross-section and centered on the long axis connecting the apex and the base is centered on the plane defined by the apex, the base and the lower part of the aortic valve 100 to 140 clockwise A plane that is rotated at a predetermined angle is specified as the cross section corresponding to the four-chamber section of the apex, and the plane is defined by the apex, the base, and the lower part of the aortic valve around the major axis. Was rotated 30 to 70 degrees counterclockwise Is defined as the cross-section corresponding to the two-chamber section of the apex, perpendicular to the plane defined by the apex, the base and the lower part of the aortic valve, and the lower part of the aortic valve and the aorta A plane passing through two points on the upper part of the valve portion is specified as the cross section corresponding to the left ventricular short axis cross section .

The present invention provides the a medical image display apparatus, the cross-section specifying unit, by using the machine learning techniques using training data, may detect the pre-Symbol feature points.

The present invention provides the a medical image display apparatus, wherein using a three-dimensional image image of the medical image, may further comprise a part identification unit that identifies a portion to be observed is included in said medical image, said cross-section a particular parts, of the 3-dimensional image of the medical image, by analyzing the 3D image image of the site are identified, among the reference cross section of one or more of the sites predetermined included in said medical image The cross section corresponding to the reference cross section of the portion to be detected may be specified.

To achieve the above object, the present invention is a computer, analyzes a medical image display method for displaying a medical image, an image obtaining step of obtaining a three-dimensional image image of the medical image, the three-dimensional image image doing, among one or more predetermined reference section, a section specifying step of specifying a cross section corresponding to the reference cross-section included in the medical image, the three-dimensional image image or al, SL before identified an image generating step of generating a cross sectional image image of the cross-sectional plane, from among the cross section specified by the section specifying step, a selection step of selecting the section indicated by the user, is generated by the image generation step, and, selecting a display control step of displaying on the display unit sectional image image of the selected the cross-section step comprises prohibits the selection of the cross-section that could not be said section specifying step for specifying the reference Refusal Includes the apex left ventricular long-axis cross-section, apex four-chamber cross-section, apex two-chamber cross-section, and left ventricular short-axis cross-section, the cross-section specifying step is to analyze the three-dimensional image of the medical image, The apex, the base, the lower part of the aortic valve, and the upper part of the aortic valve are detected as characteristic points that define the reference cross section, and the plane defined by the apex, the base and the lower part of the aortic valve is defined as the apex. Specified as the cross-section corresponding to the left ventricular long-axis cross-section, and defined by the apex, the base, and the lower part of the aortic valve centering on the long axis connecting the apex and the base A plane when the plane is rotated clockwise by 100 to 140 degrees is specified as the cross section corresponding to the cross section of the apex portion, and the apex portion, the base portion, and the aortic valve are centered on the long axis. Counterclockwise around the plane defined by the lower part A plane when rotated by 30 to 70 degrees is specified as the cross section corresponding to the apical portion two-chamber cross section, orthogonal to the plane defined by the apex portion, the base portion, and the lower part of the aortic valve portion, and A plane passing through two points of the lower part of the aortic valve part and the upper part of the aortic valve part is specified as the cross section corresponding to the left ventricular short-axis cross section .

To achieve the above object, the present invention is a program, set the computer, image acquisition means for acquiring a 3-dimensional image image of the medical image, by analyzing the 3-dimensional image image, one or more pre of the reference cross section is the cross section specifying means for specifying a cross section corresponding to the reference cross-section included in the medical image, the three-dimensional image image or al, sectional picture image of Kidan surface before said cross-section identifying unit has identified An image generation means for generating the image, a selection means for selecting the cross section instructed by the user from the cross sections specified by the cross section specifying means, and the image generation means for generating and selecting by the selection means by display control means for displaying on the display unit sectional image image of the cross section, to function as, prohibits the selection by said selecting means of said cross-section that could not be said section identifying unit identifies, The reference cross section includes the apex left ventricle long-axis cross section, apex four-chamber cross section, apex two-chamber cross section, and left ventricular short-axis cross section, and the cross-section specifying means analyzes the three-dimensional image of the medical image. Thus, the apex, base, lower part of the aortic valve, and upper part of the aortic valve are detected as characteristic points that define the reference cross section, and are defined at the apex, the base of the heart, and the lower part of the aortic valve. A plane is specified as the cross-section corresponding to the left ventricular long-axis cross section of the apex, and is defined by the apex, the base, and the lower part of the aortic valve centering on the long axis connecting the apex and the base The plane when the plane is rotated 100 to 140 degrees clockwise is specified as the cross section corresponding to the four-chamber section of the apex, and the apex, the base, and The plane defined by the lower part of the aortic valve A plane when rotated clockwise by 30 to 70 degrees is specified as the cross section corresponding to the two-chamber section of the apex and orthogonal to the plane defined by the apex, the base of the heart, and the lower part of the aortic valve In addition, a plane passing through two points of the lower part of the aortic valve part and the upper part of the aortic valve part is specified as the cross section corresponding to the left ventricular short-axis cross section .

  According to the present invention, by analyzing the three-dimensional image data, a reference cross-section included in the medical image is specified from one or more predetermined reference cross-sections, and the cross-section specifying unit is determined from the three-dimensional image data. Since the image data of the identified reference section is generated and the image of the reference section is displayed on the display unit using the generated image data, the user can easily view the image of the reference section.

  In addition, since the reference section designated by the user is selected from the specified reference sections and the image of the selected reference section is displayed on the display unit, the user can easily see the reference section to be viewed. Can do.

1 is an electrical schematic block diagram of a medical image display apparatus according to an embodiment. It is a figure which shows the image of a predetermined reference | standard cross section, when the site | part to observe is a heart. It is a flowchart which shows operation | movement of a medical image display apparatus. Three feature points on the apex-left ventricular long-axis cross section are shown. It is a figure which shows an example of positive teacher data. It is a figure which shows the feature point which is the aortic valve part upper part on the image of the apex left ventricle long-axis cross section. It is a figure which shows an example of the screen displayed on the display part by operation | movement of step S8. It is a figure which shows the other example of the screen displayed on the display part by operation | movement of step S8.

  The medical image display method according to the present invention will be described in detail below with reference to the accompanying drawings, showing preferred embodiments in relation to a medical image display apparatus and a program for executing the medical image display method.

  FIG. 1 is an electrical schematic block diagram of a medical image display apparatus 10 according to an embodiment. The medical image display device 10 includes an image acquisition unit 12, a region specifying unit 14, a cross-section specifying unit 16, an image generating unit 18, a display control unit 20, a display unit 22, and a selection unit 24. By causing the information processing apparatus (computer) to read a predetermined program, the information processing apparatus serves as the image acquisition unit 12, the region specifying unit 14, the cross-section specifying unit 16, the image generating unit 18, the display control unit 20, and the selecting unit 24. Function.

  The image acquisition unit 12 acquires 3D image data of a medical image. The medical image is an image captured by an imaging device that captures a medical image such as an X-ray imaging apparatus, an X-ray CT apparatus, an ultrasonic diagnostic apparatus, or an MRI apparatus. The image acquisition unit 12 outputs the acquired three-dimensional image data of the medical image to the part specifying unit 14.

  The site specifying unit 14 specifies a site to be observed included in the acquired medical image (for example, heart, lung, stomach, etc.). When it is difficult to directly specify a site to be observed, the site specifying unit 14 detects a site (reference site) different from the site to be observed included in the medical image, and based on the position of the detected reference site. It is also possible to indirectly specify the region of the part to be observed. For example, when the part to be observed is the heart, the part specifying unit 14 may detect the aorta and specify the heart region based on the detected position of the aorta. Further, the part specifying unit 14 may directly recognize and specify a part to be observed.

  The cross-section specifying unit 16 analyzes the three-dimensional image data of the part specified by the part specifying unit 14 out of the three-dimensional image data of the medical image, so that among the reference cross sections of the one or more predetermined parts, A reference cross section of the part included in the acquired medical image is specified. Specifically, the cross-section specifying unit 16 analyzes the three-dimensional image data of the medical image, and detects two or more feature points that define the reference cross-section of the site to be observed, thereby including the medical image in the medical image. A reference cross section of the portion to be identified is specified. The reference cross section is, for example, a cross section desired by a doctor such as a cross section defined anatomically.

  The cross section specifying unit 16 outputs three-dimensional image data of the medical image and information indicating the position of the specified reference cross section to the image generating unit 18. The information indicating the position of the reference cross section is information for specifying where the specified reference cross section exists in the medical image. Further, the cross-section specifying unit 16 may output information indicating the specified reference cross-section to the display control unit 20. Further, the cross-section specifying unit 16 may output information indicating the region of the specified part acquired from the part specifying unit 14 to the image generating unit 18.

  The image generation unit 18 generates image data of the reference cross section specified by the cross section specifying unit 16 from the three-dimensional image data of the medical image. The generation of the image data of the reference cross section is performed using information indicating the position of the reference cross section. The image generation unit 18 outputs the generated image data of the reference cross section and information indicating the generated reference cross section to the display control unit 20. The image generation unit 18 may generate image data of the reference cross section from the three-dimensional image data of the specified part among the three-dimensional image data of the medical image.

  The display control unit 20 causes the display unit 22 to display an image of the reference cross section using the image data of the reference cross section generated by the image generation unit 18. The display unit 22 is a device that can display an image, such as a cathode ray tube, a liquid crystal display, a plasma display, or a video projector.

  The selection unit 24 selects a reference cross section designated by the user from the reference cross sections specified by the cross section specifying unit 16. The user can instruct any reference cross section by operating an input device (not shown) such as a mouse, a touch panel, or a keyboard. The selection unit 24 outputs information indicating the selected reference cross section to the image generation unit 18 or the display control unit 20.

  FIG. 2 is a diagram showing an image of a predetermined reference cross section when the site to be observed is the heart. An image 50 shows an image of the apex left ventricular long-axis cross section, and an image 52 shows an image of the apex four-chamber cross section. Moreover, the image 54 has shown the image of the cardiac apex part two-chamber cross section, and the image 56 has shown the image of the left ventricular short-axis cross section. When the site to be observed is an organ other than the heart (for example, stomach, lung, etc.), the predetermined reference cross section is a reference cross section relating to the organ other than the heart.

  Next, the operation of the medical image display apparatus 10 will be described with reference to the flowchart of FIG. Here, the heart is taken as an example of the site to be observed. The image acquisition unit 12 acquires three-dimensional image data of a medical image (Step S1). The image acquisition unit 12 acquires 3D image data of a medical image from an external device. The external device may be a photographing device that captures a medical image, or may be a device that records a medical image.

  Next, the part specifying unit 14 detects the aorta included in the medical image in order to specify the heart (step S2). The part specifying unit 14 detects the aorta using the acquired three-dimensional image data of the medical image. The part specifying unit 14 determines whether or not an aorta is included in a medical image by using a classifier calculated using AdaBoost, which is one of machine learning methods. Data representing the aorta is used as positive teacher data, and tissues other than the aorta are randomly prepared as negative teacher data. The feature amount uses a combination of values of n pixel pairs selected at random. For each axial cross-sectional image of the input 3D medical image data, it is determined whether or not an aorta exists, and if it exists, the position of the center of gravity of the score is stored as a detection position. When the aorta is not detected, the part specifying unit 14 determines that the acquired medical image does not include the heart. The site specifying unit 14 may detect the aorta by other methods such as a template matching method and a demarcation method.

  Next, the part specifying unit 14 specifies a heart region in the medical image based on the detected position of the aorta (step S3). Since the aorta and the heart are in a predetermined positional relationship, if the aorta can be detected, the heart region can be specified.

  Next, the cross-section specifying unit 16 detects a predetermined feature point in the heart (step S4). The cross-section specifying unit 16 detects two or more feature points that specify the reference cross-section. FIG. 4 shows three feature points on the apical left ventricular long-axis cross section. The feature point 60 indicates the apex of the heart, the feature point 62 indicates the base of the heart, and the feature point 64 indicates the lower part of the aortic valve. The cross-section specifying unit 16 detects these three feature points 60, 62, and 64.

  The cross-section specifying unit 16 determines whether or not the feature points 60, 62, and 64 are included in the medical image by using a discriminator calculated using AdaBoost, which is one of machine learning methods. FIG. 5 is a diagram illustrating an example of positive teacher data. Images 70, 72, and 74 shown in FIG. 5 are coronal cross-sectional images of the heart. The image in the frame 80 of the image 70 shows the positive teacher data of the base of the heart, the image in the frame 82 of the image 72 shows the positive teacher data of the apex of the heart, and is in the frame 84 of the image 74. These images show positive teacher data at the lower part of the aortic valve.

  The cross-section specifying unit 16 scans the coronal cross-sectional image of the heart by using a discriminator calculated using AdaBoost, and the center of gravity of the score obtained as a result is the position of the feature points 60, 62, and 64. And When obtaining the centroid position of the score, the labeling process may be performed once and the centroid position of the label with the largest volume may be used.

  Next, the cross-section specifying unit 16 specifies a reference cross-section based on the detected feature points (step S5). The cross-section specifying unit 16 specifies the plane defined by the three feature points 60, 62, and 64 as the apex left ventricular long-axis cross section. The cross-section specifying unit 16 defines a line connecting the feature point 60 of the apex and the feature point 62 of the heart as the major axis of the heart, and the three feature points 60, 62, The plane when the plane defined by 64 is rotated 100 to 140 degrees clockwise is specified as the apex four-chamber cross section. Here, the clockwise direction in the present embodiment is the clockwise direction when viewed from the apex to the base.

  In addition, the cross-section specifying unit 16 sets the plane when the plane defined by the three feature points 60, 62, and 64 is rotated 30 to 70 degrees counterclockwise around the long axis of the heart. Identified as a cross-chamber section. The cross-section specifying unit 16 uses a plane perpendicular to the plane defined by the three feature points 60, 62, and 64 as a left ventricular short-axis cross section among two planes passing through the lower part of the aortic valve part and the upper part of the aortic valve part. Identify. The lower part of the aortic valve part is a feature point 64, and when the feature point 64 is detected, the upper part of the aortic valve part can be detected together.

  FIG. 6 is a diagram showing a feature point that is an upper part of the aortic valve portion on the cross section of the long axis of the left ventricle of the apex. A feature point 66 indicates the upper part of the aortic valve. When generating positive teacher data for determining the lower part of the aortic valve, the size and angle of the teacher data are normalized at two points, the upper part of the aortic valve part and the lower part of the aortic valve part. That is, two points are used as landmarks. Thereby, the feature points 64 and 66 can be detected.

  Next, the image generation unit 18 generates observable reference cross-section image data from the three-dimensional image data of the medical image (step S6). The observable reference cross section refers to a reference cross section specified by the cross section specifying unit 16. In step S5, since the apex left ventricular long-axis cross section, apex four-chamber cross section, apex two-chamber cross section, and left ventricular short-axis cross section are specified, the image generation unit 18 performs image data of the apex left ventricular long-axis cross section. Then, image data of the apex four-chamber section, image data of the apex two-chamber section, and image data of the left ventricular short-axis section are generated. The image generation unit 18 may generate image data of the reference cross section specified by the cross section specifying unit 16 from the three-dimensional image data of the heart specified in step S3 among the three-dimensional images of the medical image. .

  Next, the display control unit 20 displays information for allowing the user to select an observable reference cross section on the display unit 22 (step S7). The user can instruct a reference cross section to be displayed by operating the input device by looking at the screen displayed on the display unit 22 in the operation of step S7. The selection unit 24 selects a reference image instructed by the user.

  Next, the display control unit 20 causes the display unit 22 to display an image of the reference cross section selected according to the user's operation (step S8). FIG. 7 is a diagram illustrating an example of a screen displayed on the display unit 22 in the operation of step S8. A predetermined reference section name is displayed on the right side of the screen, and a check box is displayed on the left side of the name. A black check box represents the currently selected reference section. On the right side of the screen, an image of the currently selected reference cross section is displayed.

  Further, when the user operates the input device and designates a check box of a reference section to be viewed next in a state where the screen as shown in FIG. 7 is displayed, the selection unit 24 is instructed. Select the reference cross section. After that, the display control unit 20 paints the check box of the selected reference section in white, paints the chuck box of the newly selected reference section in black, and displays an image of the newly selected reference section.

  FIG. 8 is a diagram illustrating another example of the screen displayed on the display unit 22 in the operation of step S8. In FIG. 8, when there is a reference cross section that is not observable (a reference cross section that is not specified by the cross section specifying unit 16) among predetermined reference cross sections, the reference cross section cannot be selected and displayed. For example, the case where the left ventricular minor axis cross section could not be specified is taken as an example. In such a case, an X mark is put in the check box of the left ventricular minor axis section so that the left ventricular minor axis section cannot be selected. Yes. Thereby, the user can recognize that the reference cross section which cannot be selected is not included in the medical image acquired in step S1.

  In this way, by specifying the part to be observed using the three-dimensional image data of the medical image and analyzing the three-dimensional image data of the medical image, the part included in the medical image in the reference cross section of the part to be observed Since the reference cross section is specified, the specified reference cross section image data is generated from the three-dimensional image data, and the reference cross section image is displayed on the display unit, the user can easily display the reference cross section image. Can see.

  Moreover, since the reference cross section designated by the user is selected from the specified reference cross sections and an image of the selected reference cross section is displayed, the user can see the reference cross section to be viewed.

  You may deform | transform the said embodiment into the following aspects.

  (1) In the above embodiment, the apex four-chamber cross section and the apex two-chamber cross section are specified by rotating the apex left ventricular long-axis cross section in the long axis direction. And a plurality of feature points that respectively define the apex two-chamber cross section, and the apex four-chamber cross section and the apex two-chamber cross section may be directly specified based on the detected feature points.

  (2) In the above embodiment, the part to be observed included in the medical image is specified, but the part need not be specified. In this case, when the operation of step 1 is performed, the process proceeds to step S4 as it is, and the feature point may be detected. In this case, in step S4, a feature point detection process is performed for a part other than the part to be observed.

  (3) In the above embodiment, the heart is taken as an example of the site to be observed, but other sites may be used. Needless to say, the feature points to be detected change as the site changes. Further, the site to be observed may be changed according to the type of modality. Further, the reference cross section may be different depending on the type of modality.

  (4) The medical image display apparatus 10 may include a recording unit that records a sample image of a reference cross section as shown in FIG. In this case, in step S8, the display control unit 20 may display the selected reference cross-section image and simultaneously display the selected reference cross-section sample image. Thereby, a sample image and the image | photographed image of the reference | standard cross section of the patient can be compared.

  (5) The medical image display apparatus 10 may include a recording control unit that records the image data generated by the image generation unit 18 in the recording unit. In this case, the recording control unit may record the image data and also record information indicating which reference cross section the image data is. Thereby, it is possible to display an image of a reference cross section to be easily viewed later.

(6) If the aorta cannot be detected in step S2, that is, if the observation site cannot be specified, the processing may be terminated as it is. At this time, the display control unit 20 may display a message indicating that the site desired to be observed could not be specified or that the site cannot be observed. Thereby, the user can recognize that there is no part to be observed in the medical image. Further, if a 3D image of the medical image does not include a part including the heart (for example, the chest and the abdomen), it may be determined that the part to be observed cannot be specified or cannot be observed. . The site recognition of the site including the heart can be performed by site recognition described in Japanese Patent Laid-Open No. 2008-259682. Japanese Patent Application Laid-Open No. 2008-259682 performs normalization processing on a plurality of tomographic images constituting a three-dimensional image, calculates a large number of feature amounts from the normalized tomographic images, and calculates the calculated feature amounts. A score sc _np (P = head, head and neck, neck, chest, chest abdomen, abdomen, pelvis, leg) for each part representing the likelihood of the part is calculated by inputting to the discriminator obtained by the Adaboost method Then, using the calculated score sc _np as an input, the part Pn represented in the tomographic image is determined so that the arrangement order of the body parts is maintained using dynamic programming.

  Normalization processing refers to extracting a human body region from a tomographic image, calculating a landmark (reference point) from the information of the human body region, performing affine transformation with the calculated landmark as a reference, and enlarging or reducing the tomographic image. , A process of generating a normalized image by translation and rotation. This normalization process eliminates variations in the size and inclination of the human body region in the image due to individual differences, imaging conditions, etc., and aligns structures (eg, bone regions and air regions) in the human body region. The accuracy of part recognition is improved.

  The feature amount is calculated by threshold processing based on pixel values in the block (for example, 3 × 3 pixels) set in the normalized image, average values of pixel values, maximum values, minimum values, intermediate values, and CT values. The ratio of the air region and the bone region to the human body region in the human body region extracted by.

In calculating the partial score, the feature amount is input to the classifier group for each part obtained by learning based on the Adaboost method, and a part score sc _np representing the likelihood of each part is calculated for each tomographic image. The classifier group that discriminates one part includes one or more classifiers. When two or more classifiers are included, each classifier has a complementary relationship with respect to the discrimination ability. Yes.

The determination of the site is a reference site arranged in the order of the arrangement of the body part of the human body prepared in advance, that is, the head, head and neck, neck, chest, chest abdomen, abdomen, pelvis, and leg. The site Pn represented in each cross-sectional layer is finally determined so that there is no discrepancy between the maximum value of the site score sc _np in each tomographic image. Here, when there is a discrepancy between the reference part and the maximum value of the part score sc _np in each tomographic image, the part Pn is determined by obtaining a path with the minimum cost and the shortest cost so as to be costly. Final decision.

  (7) In the above embodiment, all the image data of the observable reference cross section is generated. However, after the selection unit 24 selects the reference cross section by the user operation, the image generation unit 18 selects the reference cross section. The generated image data of the reference cross section may be generated, and the display control unit 20 may display the image of the reference cross section.

  (8) In the above-described embodiment, the image acquisition unit 12 acquires the three-dimensional image data of the medical image captured by the imaging device that captures the medical image. However, the image acquisition unit 12 may perform volume rendering, MIP, minIP, or the like. You may make it acquire the produced | generated three-dimensional image. In this case, an image of the reference cross section of the three-dimensional image generated by volume rendering or the like is displayed.

  (9) The aspect which combined the said modification (1)-(8) arbitrarily may be sufficient.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

DESCRIPTION OF SYMBOLS 10 ... Medical image display apparatus 12 ... Image acquisition part 14 ... Region | part specific part 16 ... Section identification part 18 ... Image generation part 20 ... Image display control part 22 ... Display part 24 ... Selection part

Claims (5)

An image acquisition unit for acquiring a three-dimensional image of a medical image;
By analyzing the three-dimensional image, among one or more predetermined reference sections, a section specifying unit that specifies a section corresponding to the reference section included in the medical image;
An image generation unit that generates a cross-sectional image of the cross section specified by the cross-section specifying unit from the three-dimensional image;
A selection unit that selects the cross-section instructed by a user from the cross-sections specified by the cross-section specifying unit;
A display control unit that displays the cross-sectional image of the cross-section generated by the image generation unit and selected by the selection unit;
With
Prohibiting selection by the selection unit of the cross section that the cross-section specifying unit could not be specified ,
The reference cross section includes the apex left ventricular long-axis cross section, apex four-chamber cross section, apex two-chamber cross section, and left ventricular short-axis cross section,
The cross-section specifying unit detects the apex, base, aortic valve lower part, and aortic valve upper part as characteristic points that define the reference cross-section by analyzing the three-dimensional image of the medical image, A plane defined by the apex, the base of the heart, and the lower part of the aortic valve is specified as the cross section corresponding to the left ventricular long axis cross section of the apex and centered on the long axis connecting the apex and the base of the heart A plane when the plane defined by the apex, the base and the lower part of the aortic valve is rotated 100 to 140 degrees clockwise is specified as the cross section corresponding to the apical four-chamber cross section, The plane when the plane defined by the apex, the base, and the lower part of the aortic valve is rotated 30 to 70 degrees counterclockwise around the major axis is a cross section of the apex at the two chambers. Identified as the corresponding cross-section The plane perpendicular to the plane defined by the apex, the base, and the lower part of the aortic valve and passing through two points of the lower part of the aortic valve and the upper part of the aortic valve corresponds to the left ventricular short-axis cross section. A medical image display device characterized by being specified as a cross section . The medical image display device according to claim 1 ,
The sectional identification unit, by using the machine learning techniques using training data, medical image display apparatus characterized by detecting the pre-Symbol feature points. The medical image display device according to claim 1 or 2 ,
Using a three-dimensional image of the medical image, further comprising a part specifying unit for specifying a part to be observed included in the medical image;
The cross-section specifying unit analyzes the three-dimensional image of the specified part of the three-dimensional image of the medical image, so that the medical image of one or more predetermined reference cross-sections of the part is analyzed. A medical image display device that identifies the cross-section corresponding to the reference cross-section of the part included in the medical image display device. A medical image display method in which a computer displays a medical image,
An image acquisition step of acquiring a three-dimensional image of the medical image;
By analyzing the three-dimensional image, among one or more predetermined reference sections, a section specifying step for specifying a section corresponding to the reference section included in the medical image;
An image generation step of generating a cross-sectional image of the specified cross-section from the three-dimensional image;
A selection step of selecting the cross-section instructed by a user from the cross-sections specified by the cross-section specifying step;
A display control step of displaying a cross-sectional image of the cross-section generated in the image generation step and selected in the selection step on a display unit;
With
Prohibiting selection of the cross-section that could not be identified by the cross-section identification step ,
The reference cross section includes the apex left ventricular long-axis cross section, apex four-chamber cross section, apex two-chamber cross section, and left ventricular short-axis cross section,
The cross-section specifying step detects the apex, base, aortic valve lower part, and aortic valve upper part as feature points that define the reference cross-section by analyzing the three-dimensional image of the medical image, A plane defined by the apex, the base of the heart, and the lower part of the aortic valve is specified as the cross section corresponding to the left ventricular long axis cross section of the apex and centered on the long axis connecting the apex and the base of the heart A plane when the plane defined by the apex, the base and the lower part of the aortic valve is rotated 100 to 140 degrees clockwise is specified as the cross section corresponding to the apical four-chamber cross section, The plane when the plane defined by the apex, the base, and the lower part of the aortic valve is rotated 30 to 70 degrees counterclockwise around the major axis is a cross section of the apex at the two chambers. Identified as the corresponding cross-section, A plane perpendicular to the plane defined by the apex of the heart, the base of the heart, and the lower part of the aortic valve and passing through two points of the lower part of the aortic valve and the upper part of the aortic valve corresponds to the left ventricular short-axis cross section. A medical image display method characterized by specifying the cross section . Computer
Image obtaining means for obtaining three-dimensional images image of the medical image,
By analyzing the three-dimensional image, a cross-section specifying means for specifying a cross-section corresponding to the reference cross-section included in the medical image among one or more predetermined reference cross-sections,
Image generating means for generating a cross-sectional image of the cross section specified by the cross-section specifying means from the three-dimensional image;
Selecting means for selecting the cross-section instructed by the user from the cross-sections specified by the cross-section specifying means;
Display control means for causing the display section to display a cross-sectional image of the cross-section generated by the image generation means and selected by the selection means;
Function as
Prohibiting selection by the selection means of the cross section that could not be specified by the cross section specifying means ;
The reference cross section includes the apex left ventricular long-axis cross section, apex four-chamber cross section, apex two-chamber cross section, and left ventricular short-axis cross section,
The cross-section specifying means detects the apex, base, aortic valve lower part, and aortic valve upper part as characteristic points defining the reference cross-section by analyzing the three-dimensional image of the medical image, A plane defined by the apex, the base of the heart, and the lower part of the aortic valve is specified as the cross section corresponding to the left ventricular long axis cross section of the apex and centered on the long axis connecting the apex and the base of the heart A plane when the plane defined by the apex, the base and the lower part of the aortic valve is rotated 100 to 140 degrees clockwise is specified as the cross section corresponding to the apical four-chamber cross section, The plane when the plane defined by the apex, the base, and the lower part of the aortic valve is rotated 30 to 70 degrees counterclockwise around the major axis is a cross section of the apex at the two chambers. Identified as the corresponding cross-section, A plane perpendicular to the plane defined by the apex of the heart, the base of the heart, and the lower part of the aortic valve and passing through two points of the lower part of the aortic valve and the upper part of the aortic valve corresponds to the left ventricular short-axis cross section. A program characterized by being specified as the cross section .

Images