JP2006087827A - Diagnostic imaging apparatus and image processing system, program and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a diagnostic imaging apparatus capable of appropriately extracting image data in a three-dimensional region even if it is inappropriate to perform a linear interpolation perpendicular to a slice image when extracting the image data in the three-dimensional region. <P>SOLUTION: This diagnostic imaging apparatus is provided with a means 43a of detecting a position of the gravity center of a reference ROI (Region Of Interest) set in two-dimensional tomographic data, an ROI gravity center position accordance means 43b uniformly moving the two-dimensional tomographic data parallel so that the position of the gravity center of the reference ROI is positioned on a straight line perpendicular to the slice, a perpendicular linear interpolation means 43e setting the three-dimensional region by performing the linear interpolation to the two-dimensional tomographic data after the parallel movement and setting a part included in the set three-dimensional region as a region of interest, and an ROI gravity center position return means 43f moving the two-dimensional tomographic data parallel so that the position of the gravity center of the two-dimensional tomographic data after the parallel movement is returned to the position of the gravity center before the parallel movement. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image diagnostic apparatus, an image processing system, an image processing program, and an image processing method capable of extracting image data included in a three-dimensional region by interpolation of a region of interest set on a slice.

  Conventionally, it is possible to extract a three-dimensional region by interpolation of a region of interest (ROI) set on a slice from image information collected by an image diagnostic apparatus such as an X-ray CT apparatus or MRI apparatus handled in the medical field. An example of such a system is an image processing system shown in FIG. 11 (see, for example, Patent Document 1).

  As shown in FIG. 11, the conventional image processing system 1 includes an image storage unit 3, a display processing unit 5, a display unit 7, a coordinate conversion unit 9, a mouse 11, a distance image calculation unit 13, a distance image storage unit 15, and an ROI interpolation. Unit 17, ROI storage unit 19, image extraction unit 21, extracted image storage unit 23, and three-dimensional display processing unit 25.

  The image processing system 1 is connected to, for example, an X-ray CT apparatus, and the CT value for each slice obtained by the X-ray CT apparatus is stored in the image storage unit 3 as image data. The display processing unit 5 performs display processing of the image data stored in the image storage unit 3 to generate binarized image data, and gives the generated binary image data in each slice to the display unit 7. Display.

  Next, in order to extract a three-dimensional region, an outline of a reference ROI (hereinafter referred to as a reference ROI) is set by the mouse 11 in a slice image displayed on the display unit 7. The coordinate conversion unit 9 receives the reference ROI setting information from the mouse 11 and obtains the coordinates of the pixel corresponding to the outline of the reference ROI. The coordinate conversion unit 9 provides the obtained pixel coordinates to the distance image calculation unit 13.

  Next, the distance image calculation unit 13 sets the value of the pixel corresponding to the outline of the reference ROI to a predetermined value, for example, 100, and increases by 1 every time one pixel moves inside the outline of the reference ROI based on the value of the pixel. The value obtained by adding 1 to the value of the pixel and the value obtained by subtracting 1 every time the pixel advances outside the outline of the reference ROI is set to the value of the pixel. Stored in the distance image storage unit 15.

  A plurality of such reference ROIs can be selected, and distance image data is created and stored in the distance image storage unit 15 by the number of selected reference ROIs.

  Then, information for extracting a three-dimensional region by the ROI interpolation unit 17 is created from the distance image data corresponding to the reference ROI stored in the distance image storage unit 15 and written in the ROI storage unit 19.

  FIG. 12 is a diagram for explaining a three-dimensional region extraction method by the ROI interpolation unit 17 of the conventional image processing system 1 shown in FIG.

  As indicated by the solid line in FIG. 12A, ROIs on three slices (S1, S2, S3) are selected as reference ROIs (R1, R2, R3), and distance image data corresponding to each reference ROI is created. Is done. Then, the slice image interval (sampling pitch) information is input to the ROI interpolation unit 17 together with the slice image (S) indicated by the dotted line from the CT apparatus.

  Next, the ROI interpolation unit 17 performs linear interpolation on each distance image data obtained from the reference ROI. As a result, a three-dimensional region D as shown in FIG. 12B is created. Further, distance image data corresponding to each of the other slice images indicated by dotted lines is obtained from the three-dimensional region created in this way. Then, the ROI interpolation unit 17 sets an area where the pixel value of the distance image data is equal to or greater than a predetermined value as the ROI in the obtained distance image data of each slice.

  The ROI interpolation unit 17 writes the ROI created by linear interpolation and the reference ROI in the ROI storage unit 19 in association with each slice image.

  Next, when the ROI and the reference ROI of each slice image are stored in the ROI storage unit 19, the image extraction unit 21 stores them in the image storage unit 3 based on each ROI stored in the ROI storage unit 19. Of each slice image, only the ROI part or only the part outside the ROI is extracted, and is written and stored in the extracted image storage unit 23 as an extracted image.

  Next, the three-dimensional display processing unit 25 performs various three-dimensional display processes such as a superimposition process and a process for matching the gradation of the display unit 7 on the extracted image stored in the extracted image storage unit 23. Thereafter, the extracted image subjected to the three-dimensional display processing is displayed on the screen of the display unit 7 as a shading volume rendering (SVR) image, a maximum intensity projection (MIP) image, and a cross-section conversion (MPR: MPR). Multi-planar reconstruction;) is displayed as a three-dimensional image such as an image.

That is, the conventional image processing system 1 sets a reference ROI on a slice image such as a plurality of parallel MPR images as a reference, and extracts a three-dimensional region by linearly interpolating between the reference ROIs that are adjacent in position. System. Such a three-dimensional region extraction technique using linear interpolation is used as a means for extracting an organ having an arbitrary shape, for example.
JP-A-7-234927

  In the conventional image processing system 1, the positional relationship of the slice images is fixed, and linear interpolation processing is performed in the vertical direction on the fixed slice images, so that the slice images are not perpendicular to the slice image such as an organ such as a blood vessel. There is a problem that it is difficult to appropriately extract organs and organs.

  FIG. 13 is a diagram illustrating an example of a problem that occurs when the three-dimensional region extraction method by the ROI interpolation unit 17 of the conventional image processing system 1 shown in FIG. 11 is used.

  When the organ or organ to be extracted such as a blood vessel is not perpendicular to the slice image, as shown in FIG. 13A, the reference ROI (R1, R2, R3) is along the organ such as the blood vessel, that is, The slice image (S1, S2, S3) is created along the axis in the oblique direction, not along the axis in the vertical direction. As a result, the three-dimensional region D obtained by linear interpolation in the vertical direction with respect to the slice image may be a region separated for each slice image as shown in FIG.

  Another problem is that the organ or organ to be extracted has a flat cross-sectional shape that is far from the perfect circle, such as an ellipse with a large difference between the major axis and minor axis on the slice image. Even when the cross-sectional shapes are in a twisted positional relationship, extraction is difficult.

  FIG. 14 is a diagram for explaining another example of a problem that occurs when the three-dimensional region extraction method by the ROI interpolation unit 17 of the conventional image processing system 1 shown in FIG. 11 is used.

  When the cross-sectional shape of the organ to be extracted or the slice image of the organ is flat and the cross-sectional shapes are twisted between the slices, as shown in FIG. The ROIs (R1, R2) are created in a state where they are rotated relative to each other. As a result, the three-dimensional region D obtained by linear interpolation in the vertical direction with respect to the slice image (S1, S2) is a region having a sharp portion as shown in FIG.

  The present invention has been made in order to cope with such a conventional situation, and when extracting image data of a three-dimensional region, a cross section on the slice image of the three-dimensional region does not follow an axis perpendicular to the slice image. Even when it is inappropriate to perform linear interpolation in the vertical direction on the slice image, as in the case where the cross-sectional shapes are twisted between the slices, the cubic is appropriately An object of the present invention is to provide an image diagnostic apparatus, an image processing system, an image processing program, and an image processing method capable of extracting image data of an original area.

  In order to achieve the above object, an image diagnostic apparatus according to the present invention includes, as described in claim 1, image data collecting means for collecting two-dimensional tomographic image data of a subject, and the two-dimensional tomographic image data. Based on the ROI centroid position detection means for detecting the position of the centroid of the reference ROI, and the position of the centroid detected by the ROI centroid position detection means , Together with the two-dimensional tomographic image data in which the reference ROI is set so that the position of the center of gravity of the reference ROI adjacent in the slice direction of the two-dimensional tomographic image data is on a straight line in the direction perpendicular to the slice ROI centroid position matching means for evenly translating the two-dimensional tomographic image data for which no reference ROI is set in a direction parallel to the slice; and A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the parallel movement by the OI centroid position matching means, and the parallel included in the set three-dimensional region. Vertical linear interpolation means for setting a portion of the two-dimensional tomographic image data after movement as a region of interest of the two-dimensional tomographic image data, and a center of gravity of the two-dimensional tomographic image data after the parallel movement in which the region of interest is set ROI center-of-gravity position returning means for translating the two-dimensional tomographic image data so that the position of the center of gravity is returned to the position of the center of gravity before the parallel movement.

  In order to achieve the above-mentioned object, an image diagnostic apparatus according to the present invention includes an image data collection unit that collects two-dimensional tomographic image data of a subject, and the two-dimensional tomographic image as described in claim 2. Based on the ROI rotation angle detection means for detecting the rotation angle of the reference ROI with the plurality of regions of interest serving as the reference set in the image data as the reference ROI, and the rotation angle detected by the ROI rotation angle detection means, The two-dimensional tomographic image in which the reference ROI is not set together with the two-dimensional tomographic image data in which the reference ROI is set so that the rotation angles of the reference ROIs adjacent in the slice direction of the two-dimensional tomographic image data match. ROI rotation angle matching means for rotating data evenly on a plane parallel to the slice, and rotation by the ROI rotation angle matching means. A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after movement, and the two-dimensional tomogram after the rotational movement included in the set three-dimensional region Vertical linear interpolation means for setting a portion of image data as a region of interest of the two-dimensional tomographic image data, and a rotation angle of the two-dimensional tomographic image data after the rotational movement in which the region of interest is set is a rotation before the rotational movement. ROI rotation angle return means for rotating and moving the two-dimensional tomographic image data so as to return to an angle.

  In order to achieve the above object, an image processing system according to the present invention uses a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data of a subject as a reference. The ROI is adjacent to the slice direction of the two-dimensional tomographic image data based on the ROI centroid position detecting means for detecting the position of the centroid of the reference ROI and the position of the centroid detected by the ROI centroid position detecting means. The two-dimensional tomographic image data in which the reference ROI is not set together with the two-dimensional tomographic image data in which the reference ROI is set so that the position of the center of gravity of the reference ROI is on a straight line perpendicular to the slice ROI center-of-gravity position matching means for uniformly translating the image in the direction parallel to the slice, and after the parallel movement by this ROI center-of-gravity position matching means A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data, and the two-dimensional tomographic image data after the translation included in the set three-dimensional region is set. Vertical linear interpolation means for setting a portion as a region of interest of the two-dimensional tomographic image data, and the position of the center of gravity of the two-dimensional tomographic image data after the parallel movement in which the region of interest is set is the position of the center of gravity before the parallel movement ROI center-of-gravity position return means for translating the two-dimensional tomographic image data so as to return to the position.

  In order to achieve the above-mentioned object, the image processing system according to the present invention uses a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data of a subject as described in claim 4. As ROI, ROI rotation angle detection means for detecting the rotation angle of the reference ROI, and the reference ROI adjacent in the slice direction of the two-dimensional tomographic image data based on the rotation angle detected by the ROI rotation angle detection means The two-dimensional tomographic image data for which the reference ROI is not set together with the two-dimensional tomographic image data for which the reference ROI is set so that the rotation angles coincide with each other are rotated evenly on a plane parallel to the slice. ROI rotation angle matching means and the slice of the two-dimensional tomographic image data after rotational movement by the ROI rotation angle matching means. A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the two-dimensional tomographic image data after the rotational movement included in the set three-dimensional region Vertical linear interpolation means for setting the region of interest, and the two-dimensional tomographic image so that the rotation angle of the two-dimensional tomographic image data after the rotational movement in which the region of interest is set is returned to the rotation angle before the rotational movement. ROI rotation angle return means for rotating and moving data.

  In order to achieve the above-mentioned object, the image processing program according to the present invention, as set forth in claim 5, makes a computer a plurality of interests that serve as a reference set in the two-dimensional tomographic image data of the subject. Using the region as a reference ROI, ROI centroid position detection means for detecting the position of the centroid of the reference ROI, and based on the position of the centroid detected by the ROI centroid position detection means, in the slice direction of the two-dimensional tomographic image data The two-dimensional tomography in which the reference ROI is not set together with the two-dimensional tomographic image data in which the reference ROI is set so that the position of the center of gravity of the adjacent reference ROI is on a straight line perpendicular to the slice ROI barycentric position matching means for evenly translating image data in a direction parallel to the slice, and parallel by the ROI barycentric position matching means A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after movement, and the two-dimensional tomogram after the parallel movement included in the set three-dimensional region Vertical linear interpolation means for setting a portion of image data as a region of interest of the two-dimensional tomographic image data, and the position of the center of gravity of the two-dimensional tomographic image data after the parallel movement in which the region of interest is set is the center of gravity before the parallel movement It functions as ROI barycentric position returning means for translating the two-dimensional tomographic image data so that it is returned to the position.

  In order to achieve the above object, an image processing program according to the present invention provides a computer with a plurality of interests serving as a reference set in two-dimensional tomographic image data of a subject. ROI rotation angle detection means for detecting the rotation angle of the reference ROI with the region as the reference ROI, and the adjacent to the slice direction of the two-dimensional tomographic image data based on the rotation angle detected by the ROI rotation angle detection means The two-dimensional tomographic image data for which the reference ROI is not set together with the two-dimensional tomographic image data for which the reference ROI is set so that the rotation angles of the reference ROI coincide with each other on a plane parallel to the slice. ROI rotation angle matching means to be moved, and the two-dimensional tomographic image data after rotational movement by this ROI rotation angle matching means A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the rotational movement included in the set three-dimensional region. Vertical linear interpolation means for setting the region of interest of image data, and the two-dimensional so that the rotation angle of the two-dimensional tomographic image data after the rotational movement in which the region of interest is set is returned to the rotation angle before the rotational movement. It is characterized by functioning as ROI rotation angle return means for rotationally moving tomographic image data.

  In order to achieve the above object, the image processing method according to the present invention uses a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data of the subject as a reference. As the ROI, the step of detecting the position of the center of gravity of the reference ROI and the position of the center of gravity of the reference ROI adjacent in the slice direction of the two-dimensional tomographic image data based on the detected position of the center of gravity in the slice. The two-dimensional tomographic image data for which the reference ROI is not set and the two-dimensional tomographic image data for which the reference ROI is not set are equally parallel to a direction parallel to the slice so as to be on a straight line in a vertical direction. A three-dimensional region is formed by performing a linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the parallel movement. And setting a portion of the two-dimensional tomographic image data after the translation included in the set three-dimensional region as a region of interest of the two-dimensional tomographic image data, and the region of interest set Translating the two-dimensional tomographic image data so that the position of the center of gravity of the two-dimensional tomographic image data after translation is restored to the position of the center of gravity before translation. .

  In order to achieve the above-mentioned object, the image processing method according to the present invention is based on a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data of the subject. As the ROI, the step of detecting the rotation angle of the reference ROI and, based on the detected rotation angle, the rotation angle of the reference ROI adjacent in the slice direction of the two-dimensional tomographic image data match. Rotating the two-dimensional tomographic image data for which the reference ROI is not set together with the two-dimensional tomographic image data for which the reference is set on the plane parallel to the slice; and the two-dimensional after the rotational movement A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of tomographic image data, and the set three-dimensional A portion of the two-dimensional tomographic image data after the rotational movement included in a region is set as a region of interest of the two-dimensional tomographic image data; and the two-dimensional tomographic image after the rotational movement in which the region of interest is set And rotating the two-dimensional tomographic image data so that the rotation angle of the data is restored to the rotation angle before the rotation movement.

  In the diagnostic imaging apparatus, the image processing system, the image processing program, and the image processing method according to the present invention, when extracting the image data of the three-dimensional region, the cross-section on the slice image of the three-dimensional region is the axis line perpendicular to the slice image. Even if it is inappropriate to perform linear interpolation in the vertical direction on the slice image, such as when the cross-sectional shape is twisted between each slice The image data of the three-dimensional area can be appropriately extracted.

  Embodiments of an image diagnostic apparatus, an image processing system, an image processing program, and an image processing method according to the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a functional block diagram showing an embodiment of an ultrasonic diagnostic apparatus as an example of an image diagnostic apparatus according to the present invention.

  An ultrasonic diagnostic apparatus 30 as an example of an image diagnostic apparatus includes a transmission circuit 31, a reception circuit 32, an ultrasonic probe 33, an image generation circuit 34, and an image processing system 35 including a pulsar. The transmission circuit 31 and the reception circuit 32 are connected to the ultrasonic probe 33, and an electric pulse generated in the transmission circuit 31 is applied to a plurality of ultrasonic transducer groups provided in the ultrasonic probe 33, thereby not being illustrated. While the ultrasonic pulse is transmitted toward the object, the reflected wave signal generated in the subject is received by the ultrasonic probe 33 and output to the receiving circuit 32.

  The receiving circuit 32 has a function of generating an image signal by converting a reflected wave signal received from the ultrasonic probe 33 into a digital signal and performing various processes such as delay addition processing, detection, and compression.

  The image generation circuit 34 receives an image signal from the reception circuit 32 and generates image data in each slice of the subject, and gives the image data in each generated slice to the image processing system 35 as slice image data. It has a function.

  That is, the ultrasound diagnostic apparatus 30 is provided with a function as an image data collection unit that collects two-dimensional tomographic image data of a subject by the transmission circuit 31, the reception circuit 32, the ultrasound probe 33, and the image generation circuit 34. .

  The image processing system 35 includes an image storage unit 36, a display processing unit 37, a display unit 38, a coordinate conversion unit 39, a mouse 40, a distance image calculation unit 41, a distance image storage unit 42, an ROI interpolation unit 43, an ROI storage unit 44, An image extraction unit 45, an extracted image storage unit 46, and a three-dimensional display processing unit 47 are provided. The image processing system 35 can be constructed by causing a computer to read an image processing program, but part or all of the image processing system 35 may be configured by a circuit.

  The image processing system 35 is built in, for example, an ultrasonic diagnostic apparatus 30 as an example of an image diagnostic apparatus. However, the image processing system 35 may be incorporated in an image diagnostic apparatus such as another MRI apparatus or an X-ray CT apparatus, or may be configured independently of the image diagnostic apparatus and input necessary data from the image diagnostic apparatus. Information processing may be performed.

  The image storage unit 36 of the image processing system 35 stores slice image data obtained by the image generation circuit 34 in association with a plurality of two-dimensional vertical and horizontal pixels.

  The display processing unit 37 has a function of performing display processing such as processing for matching the gradation of the display unit 38 and binarization processing on the slice image data stored in the image storage unit 36. The display unit 38 includes, for example, a display device such as a CRT or a liquid crystal display device and an input unit (not shown) such as a keyboard. The display unit 38 has a function of selecting slice image data for setting a reference ROI (hereinafter referred to as a reference ROI) by the operation of the input unit, and display processing and binarization processing are performed by the display processing unit 37. And a function of displaying a slice image and an image that has been three-dimensionally displayed by a three-dimensional display processing unit 47 described later.

  The coordinate conversion unit 39 has a function of setting the reference ROI of the slice image data selected on the display unit 38 based on the setting information of the reference ROI input by input means such as the mouse 40, and the set reference ROI. And a function for obtaining the coordinates of the pixel corresponding to the contour. As the reference ROI setting information, for example, the amount of movement of the mouse 40 can be used.

  The distance image calculation unit 41 sets the pixel value corresponding to the coordinates of the contour of the reference ROI obtained by the coordinate conversion unit 39 to a predetermined value, and sets one pixel inside the contour of the reference ROI based on the value of the pixel. Each time a value is added, a value obtained by adding 1 is used as the value of the pixel, and a value obtained by subtracting 1 every time the pixel advances outside the outline of the reference ROI is used as the value of the pixel. It has a function for obtaining all pixels as image data.

  The distance image storage unit 42 stores the distance image data obtained by the distance image calculation unit 41.

  The ROI interpolation unit 43 reads the distance image data having the reference ROI from the distance image storage unit 42, receives slice image data and slice image interval (sampling pitch) information from the image generation circuit 34, and performs geometric coordinate conversion. By performing linear interpolation using the reference ROI together with the processing, a function for obtaining distance image data corresponding to each slice image data, and a region where the pixel value of each obtained distance image data is a predetermined value or more is set as the ROI It has the function to do.

  FIG. 2 is a functional block diagram showing a detailed configuration of the ROI interpolation unit 43 of the image processing system 35 built in the ultrasonic diagnostic apparatus 30 shown in FIG.

  The ROI interpolation unit 43 includes an ROI centroid position detection unit 43a, an ROI centroid position matching unit 43b, an ROI rotation angle detection unit 43c, an ROI rotation angle matching unit 43d, a vertical linear interpolation unit 43e, an ROI centroid position return unit 43f, and an ROI rotation angle. The return part 43g has these components, and the ROI interpolation part 43 has the ROI center of gravity position detecting means, the ROI center of gravity position matching means, the ROI rotation angle detection means, the ROI rotation angle matching means, the vertical linear interpolation means, respectively. Functions as ROI barycentric position return means and ROI rotation angle return means are provided.

  The ROI barycentric position detection unit 43a has a function of detecting the position of the barycenter of each reference ROI in each distance image data read from the distance image storage unit 42, and the position of the center of gravity of each detected reference ROI as ROI. It has a function of giving to the center-of-gravity position matching part 43b and the ROI center-of-gravity position return part 43f. However, the input data of the ROI barycentric position detecting unit 43a may be data other than the distance image data such as data in which the ROI inside is a constant value and the outside of the ROI is 0.

  FIG. 3 is a diagram for explaining an example of input data of the image processing system 35 built in the ultrasonic diagnostic apparatus 30 shown in FIG.

  3A-1 is distance image data that is an example of input data of the ROI barycentric position detection unit 43a, and FIG. 3A-2 is a pixel value in the dotted line part of FIG. 3A-1. This is a histogram.

  As shown in FIGS. 3A-1 and 3A-2, the reference ROI of the distance image data is defined as an area having a pixel value of 100 or more, for example.

  FIG. 3B-1 is data in which the ROI inside the ROI, which is an example of the input data of the ROI barycentric position detector 43a, has a constant value 1 and 0 outside the ROI. FIG. It is a histogram of the pixel value in the dotted line part of b-1).

  As shown in FIGS. 3 (b-1) and 3 (b-2), when data having a constant value of 1 within the ROI and 0 outside the ROI is used as input data to the ROI barycentric position detection unit 43a, The ROI is defined as an area having a value of 1, for example.

  On the other hand, the ROI center-of-gravity position matching unit 43b receives each slice image data for which the reference ROI is not set from the image generation circuit 34, and receives the positional deviation information of the center of gravity of each reference ROI from the ROI center-of-gravity position detection unit 43a. Each slice image data in which the reference ROI is not set is sliced together with the distance image data in which the reference ROI is set so that the center of gravity position of each reference ROI adjacent in the slice direction is on a straight line in the direction perpendicular to the slice. A function of evenly translating in parallel directions, and a function of supplying each distance image data and slice image data after translation to the ROI rotation angle matching unit 43d or the vertical linear interpolation unit 43e.

  The ROI rotation angle detection unit 43c has a function of detecting the rotation angle of each reference ROI in each distance image data read from the distance image storage unit 42 by an arbitrary method such as the central moment method, and the rotation of each detected reference ROI. A function of giving the angle to the ROI rotation angle matching unit 43d and the ROI rotation angle return unit 43g as angle deviation information. The input data of the ROI rotation angle detection unit 43c may be data other than the distance image data such as data in which the ROI inside is a constant value and the outside of the ROI is 0.

  Here, the rotation angle of the reference ROI can be defined by an arbitrary method. For example, it is also possible to set the inclination of a vector whose component is an orthogonal projection of the reference ROI to the coordinate axis with respect to the coordinate axis. In the case of the central moment method, the rotation angle of the reference ROI is an angle formed by the direction in which the figure of the reference ROI extends and the coordinate axis.

  The calculation formula of the rotation angle θ of the reference ROI by the central moment method is expressed by the formula (1).

[Equation 1]
θ = (1/2) tan ⁻¹ {2M ₁₁ / (M ₂₀ −M ₀₂ )} (1)

However, in the formula (1), the central moments M ₁₁ , M ₂₀ , M ₀₂ are the function f (i, j) representing the contour of the reference ROI and the barycentric coordinates (i _g , j _g ) of the function f (i, j). ) From the above equation (2).

[Equation 2]
Mpq = Σ _i Σ _j (i−i _g ) ^p · (j−j _g ) ^q · f (i, j) (2)

  The ROI rotation angle matching unit 43d receives the distance image data and the slice after translation received from the ROI barycenter position matching unit 43b based on the angular deviation information of the rotation angle of each reference ROI received from the ROI rotation angle detection unit 43c. The image data is rotated and moved evenly on a plane parallel to the slice direction to match the rotation angles of the distance image data and the slice image data, and the distance image data and the slice image data after the rotation are moved. And a function to be given to the vertical linear interpolation unit 43e.

  The vertical linear interpolation unit 43e uses the distance image data and the slice image data after the parallel movement or the rotational movement received from the ROI center-of-gravity position matching unit 43b or the ROI rotation angle matching unit 43d. Included in the set three-dimensional region of the slice image data after parallel movement or rotational movement, and the function of setting the three-dimensional region (Voxel data) by performing linear interpolation in the direction perpendicular to the slice of the image data And a function of setting the portion as the ROI of each slice image data.

  Then, the vertical linear interpolation unit 43e uses the reference ROI after movement and the ROI of each set slice image data as ROI information, along with each distance image data and slice image data after translation or rotation, and ROI barycenter position return unit 43f or ROI rotation angle return unit 43g.

  The ROI rotation angle return unit 43g, based on the angular position shift information of the rotation angle of the reference ROI received from the ROI rotation angle detection unit 43c, the rotation angle of each distance image data and slice image data received from the vertical linear interpolation unit 43e. Has a function of rotating the ROI rotation angle matching unit 43d so as to return to the rotation angle before the rotation movement, and a function of giving each ROI image data and slice image data to the ROI barycentric position return unit 43f after the rotation angle is returned. Have

  The ROI center-of-gravity position return unit 43f receives each distance image data and slice image received from the vertical linear interpolation unit 43e or the ROI rotation angle return unit 43g based on the positional deviation information of the reference ROI received from the ROI center-of-gravity position detection unit 43a. A function of moving the center of gravity of the data so that the center of gravity of the data is returned to the position of the center of gravity before the parallel movement by the ROI center of gravity position matching unit 43b, and the ROI information together with each distance image data and slice image data after the return of the center of gravity position 44 has a function of writing.

  Therefore, the ROI storage unit 44 of the image processing system 35 stores the ROI set by the ROI interpolation unit 43 and the reference ROI in association with each distance image data and slice image data.

  The image extraction unit 45 is based on each ROI and reference ROI stored in the ROI storage unit 44, and includes each slice image data stored in the image storage unit 36 as original data, and is included in each ROI and reference ROI. A function for generating extracted image data within the ROI by performing the data extraction process, a function for generating image data extracted outside the ROI composed of image data outside the ROI, and other image processing such as paint processing as necessary Has a function of generating a paint image.

  The extracted image storage unit 46 stores in-ROI extracted image data, out-ROI extracted image data, and paint image data generated by the ROI extracting unit 21.

  The three-dimensional display processing unit 47 binarizes, superimposes, and displays the level of the display unit 38 on the extracted image data in the ROI, the extracted image data outside the ROI, and the paint image data stored in the extracted image storage unit 46. It has a function of generating 3D image data for displaying a 3D image such as an SVR image, a MIP image, or an MPR image by performing a 3D display process such as a process for adjusting the key.

  The image storage unit 36, the distance image storage unit 42, the ROI storage unit 17, and the extracted image storage unit 46 are each composed of a storage device such as a video memory, for example, but may be shared by a single or a plurality of storage devices. May be. In addition, the display processing unit 37, the distance image calculation unit 41, the ROI interpolation unit 43, the image extraction unit 45, and the 3D display processing unit 47, for example, read various programs into an arithmetic processing device such as a CPU or a microprocessor. Although configured, a single or a plurality of arithmetic processing devices may be shared.

  Next, the operation of the ultrasonic diagnostic apparatus 30 will be described.

  FIG. 4 is a flowchart showing a procedure for displaying an image included in a three-dimensional region by the ultrasonic diagnostic apparatus 30 shown in FIG. 1, and the reference numerals with numerals in the figure indicate the steps of the flowchart. .

  First, in step S1, slice image data of the subject is collected. That is, the electric pulse generated in the transmission circuit 31 is applied to a plurality of ultrasonic transducer groups provided in the ultrasonic probe 33, and the ultrasonic pulse is transmitted toward a subject (not shown). Then, the reflected wave signal generated in the subject is received by the ultrasonic probe 33 and output to the receiving circuit 32.

  Further, the reception circuit 32 converts the reflected wave signal received from the ultrasonic probe 33 into a digital signal, and then performs various processes such as delay addition processing, detection, and compression to generate an image signal. The image generation circuit 34 receives the image signal from the reception circuit 32 and generates image data in each slice of the subject as slice image data. The generated slice image data is written and stored in the image storage unit 36 of the image processing system 35.

  When the blood vessel portion data is extracted from the slice image data collected in this way and the blood vessels are displayed as a three-dimensional image such as an SVR image, MIP image, or MPR image, they are displayed. It is necessary to set a three-dimensional region that includes blood vessels. Therefore, various image processing for setting a three-dimensional region and displaying a blood vessel as a three-dimensional image by the image processing system 35 is performed.

  That is, in step S2, slice image data is selected in order to set a reference ROI serving as a reference for setting a three-dimensional region on the slice image data. For this purpose, the display processing unit 37 performs display processing of the image data stored in the image storage unit 36 to generate binarized image data, and displays the generated binary image data in each slice on the display unit 38. Display by giving. Therefore, the user can select slice image data for setting the reference ROI by operating an input device (not shown) provided on the display unit 38.

  Next, in step S3, the outline of the reference ROI is set by the input means such as the mouse 40 in the selected slice image data. That is, the coordinate conversion unit 39 receives the reference ROI setting information from the mouse 40 and obtains the coordinates of the pixel corresponding to the outline of the reference ROI. Then, the coordinate conversion unit 39 gives the obtained pixel coordinates to the distance image calculation unit 41.

  Next, in step S <b> 4, distance image data is created by the distance image calculation unit 41 and stored in the distance image storage unit 42. That is, the distance image calculation unit 41 sets the value of the pixel corresponding to the contour of the reference ROI to a predetermined value, for example, 100, and sets 1 for every one pixel inward of the contour of the reference ROI based on the value of the pixel. The added value is used as the value of the pixel, and the value obtained by subtracting 1 every time one pixel moves outside the outline of the reference ROI is used as the value of the pixel. Store in the image storage unit 42.

  That is, the process for using the reference ROI for setting the three-dimensional region is completed by storing the distance image data in the distance image storage unit 42. Such setting of the reference ROI and creation of the distance image data are performed a plurality of times according to the accuracy required for the three-dimensional region. That is, if more reference ROIs are set, the accuracy of the three-dimensional region can be improved.

  Therefore, if further setting of the reference ROI is necessary in step S5, slice image data is selected again in step S2, and the reference ROI is set and distance image data is created in the same procedure.

  On the other hand, when the setting of the reference ROI is completed in step S5, in step S6, the ROI interpolation unit 43 performs linear interpolation using the reference ROI together with the coordinate conversion processing of the geometric slice image data and the distance image data. A three-dimensional area is set, and a portion of each slice image data included in the set three-dimensional area is set as an ROI.

  Here, the details of the coordinate conversion process and the linear interpolation process performed by the ROI interpolation unit 43 will be described. The ROI interpolation unit 43 automatically sets the ROI for each slice image data for which the reference ROI is not set by performing linear interpolation in the direction perpendicular to the slice using the reference ROI. However, for example, when generating a three-dimensional image of a blood vessel, the desired ROI is not always along a straight line perpendicular to the slice, and the desired ROIs are in a twisted relationship between slices. There can be.

  For this reason, there is a possibility that a desired ROI may not be appropriately set by simply performing linear interpolation in the direction perpendicular to the slice using the reference ROI. Therefore, in order to avoid such a problem, the ROI interpolation unit 43 performs coordinate conversion processing of the reference ROI and each slice image data before and after the linear interpolation processing.

  5 is a flowchart showing details of a procedure for setting the ROI of each slice image data by the ROI interpolation unit 43 in the flowchart shown in FIG. 4, and FIG. 6 shows each slice image by the ROI interpolation unit 43 shown in FIG. It is a figure explaining the setting method of ROI of data.

  Moreover, the code | symbol which attached | subjected the number to S in FIG. 5 shows each step of a flowchart.

  First, in step S10, the center of gravity of each reference ROI is detected by the ROI center-of-gravity position detecting unit 43a, and the position of the center of gravity of each detected reference ROI is used as positional deviation information, and the ROI center-of-gravity position matching unit 43b and the ROI center-of-gravity position returning unit 43f. Given to. That is, as shown in FIG. 6A, for example, the center of gravity position of each reference ROI (R1, R2) on the two distance image data (S1, S2) indicated by the solid line is detected by the ROI center of gravity position detector 43a. The

  Next, in step S11, the ROI center-of-gravity position matching unit 43b receives each slice image data for which the reference ROI is not set from the image generation circuit 34, and the center-of-gravity position shift of each reference ROI from the ROI center-of-gravity position detection unit 43a. Each slice in which the reference ROI is set together with the distance image data in which the reference ROI is set so that the center of gravity of each reference ROI adjacent to the slice direction is on a straight line in the direction perpendicular to the slice. The image data is evenly translated in the direction parallel to the slice.

  That is, as shown in FIG. 6B, each slice image data (S) indicated by a dotted line in which the reference ROI is not set is read from the image generation circuit 34, and each reference ROI together with the distance image data (S1, S2). The center of gravity of (R1, R2) is evenly translated so that it is on a straight line in the direction perpendicular to the slice.

  As a result, even if the desired ROI to be set does not follow a straight line perpendicular to the slice, the ROI to be set can be made to follow the vertical straight line to the slice by translation. However, when a plurality of desired ROIs are twisted between slices, an appropriate ROI can be obtained even if linear interpolation is performed in the vertical direction to the slices using the reference ROI after translation and each slice image data. There is a possibility that it cannot be set. However, even when a plurality of desired ROIs between slices have a twist relationship, the twist relationship can be eliminated by rotating the reference ROI and each slice image data.

  Accordingly, in step S12, it is determined whether or not a plurality of desired ROIs have a twist relationship between slices. That is, the ROI rotation angle detection unit 43c detects the rotation angle of each reference ROI in each distance image data read from the distance image storage unit 42 as angle deviation information by an arbitrary method such as the central moment method. Then, based on the detected angle deviation information, it can be determined whether or not the reference ROI has a twist relationship.

  For example, as shown in FIG. 6B, when the reference ROI (R1, R2) is not flat and has no twist relationship, it is determined that the rotational movement is unnecessary.

  In step S13, the distance image data and the slice image data after the parallel movement are given to the vertical linear interpolation unit 43e, and are adjacent to each other in the slice direction based on the distance image data and the slice image data after the parallel movement. A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of each distance image data. Then, a portion included in the set three-dimensional region in the slice image data after the parallel movement is set as the ROI of each slice image data.

  That is, a three-dimensional region (D) as shown by a solid line in FIG. 6C is set, and a dotted line portion of each slice image data (S) included in the three-dimensional region (D) is set as ROI (R). The

  Here, the details of the ROI setting method by linear interpolation will be described.

  FIG. 7 is a diagram illustrating a method for setting an ROI by linear interpolation using two distance images.

  As shown in FIG. 7, in the two distance images (the input image of the slice image n-1 and the input image of the slice image n), the number of pixels on the contour of the reference ROI is 100, for example. A value obtained by adding 1 each time the value advances one inside the reference ROI is used as a pixel value, while a value obtained by subtracting 1 each time the value is advanced outside the reference ROI is set as a pixel value.

  Then, linear interpolation is performed based on these distance images, and an interpolation image is generated as a new distance image from slice image data between the distance images. That is, the pixel value at the coordinates (x, y) of the distance image generated from the slice image n-1 is n-1 (x, y), and the coordinates (x, y) of the distance image generated from the slice image n are used. Assuming that the pixel value is n (x, y), the pixel value T (x, y) at the coordinates (x, y) of the new distance image is expressed by Expression (3).

[Equation 3]
T (x, y) = {m · n (x, y) + l · n−1 (x, y)} / (l + m) (3)
Here, l and m are values determined according to the distance from the original distance image to the new distance image. For example, when a new distance image is generated in the middle of two distance images, a distance image in which the average of the pixel values of the two distance images is set as 1 = m = 1 in Equation (3). Is created.

  FIG. 8 is a diagram in which the distance image and the interpolation image shown in FIG.

  In each distance image shown in FIG. 7, when the pixel value 100 is binarized using the threshold value and expressed in a three-dimensional manner, a three-dimensional region formed on the stairs is formed so that the step becomes smaller as shown in FIG. 8.

  Then, the vertical linear interpolation unit 43e uses the ROI of each slice image data set in this way and the reference ROI after translation as ROI information, and returns the ROI centroid position together with each distance image data and slice image data after translation. This is given to the portion 43f.

  Next, in step S14, the ROI center-of-gravity position return unit 43f receives each distance image data and slice image data received from the vertical linear interpolation unit 43e based on the positional deviation information of the reference ROI received from the ROI center-of-gravity position detection unit 43a. The center of gravity position is translated so as to be restored to the position of the center of gravity before the parallel movement by the ROI center of gravity position matching portion 43b.

  That is, as shown in FIG. 6D, the ROI (R) and the reference ROI (R1, R2) are translated, and the ROI is included in the three-dimensional region (D) not along the straight line in the direction perpendicular to the slice. (R) and the reference ROI (R1, R2) are set.

  On the other hand, if it is determined in step S12 that a plurality of desired ROIs are in a torsional relationship between slices, the reference ROI and each slice image data are rotationally moved. In this case, prior to the rotational movement of the reference ROI and each slice image data, the angular displacement information of the reference ROI is given from the ROI rotation angle detection unit 43c to the ROI rotation angle matching unit 43d and the ROI rotation angle return unit 43g.

  In step S15, the distance image data and the slice image data after the parallel movement are supplied from the ROI barycentric position matching unit 43b to the ROI rotation angle matching unit 43d, and the distance image data and the slice image data are rotationally moved. It is.

  FIG. 9 is a diagram for explaining a ROI setting method when the ROI interpolation unit 43 shown in FIG. 2 rotates the reference ROI and each slice image data.

  For example, as shown in FIG. 9A, the flat reference ROI (R1, R2) is rotated by a certain angle on the two distance image data (S1, S2) by the parallel movement of the reference ROI (R1, R2). It becomes a state. Then, as shown in FIG. 9B, the ROI rotation angle coincidence unit 43d performs parallel movement as shown in FIG. 9B based on the angular deviation information of the rotation angle of each reference ROI (R1, R2) received from the ROI rotation angle detection unit 43c. The distance image data (S1, S2) and the slice image data (S) are rotated evenly on a plane parallel to the slice direction to thereby obtain the distance image data (S1, S2) and the slice image data (S). ) To match the rotation angle. Further, the ROI rotation angle matching unit 43d gives the distance image data (S1, S2) and the slice image data (S) after the rotational movement to the vertical linear interpolation unit 43e.

  Next, in step S16 of FIG. 5, the vertical linear interpolation unit 43e performs reference of each distance image data (S1, S2) after the rotational movement received from the ROI rotation angle matching unit 43d as shown in FIG. 9C. Based on the ROI (R1, R2), the three-dimensional region (D) is set by performing linear interpolation in the direction perpendicular to the slice of each distance image data (S1, S2) adjacent in the slice direction. Then, a portion included in the set three-dimensional region (D) in the slice image data (S) after the rotational movement is set as the ROI (R) of each slice image data (S). Further, the vertical linear interpolation unit 43e uses the reference ROI (R1, R2) and the ROI (R) of each set slice image data (S) as ROI information, and each distance image data (S1 after translation and rotation). , S2) and slice image data (S) are given to the ROI rotation angle return unit 43g.

  Next, in step S17, the ROI rotation angle return unit 43g is shown in FIG. 9D based on the angular position deviation information of the rotation angle of the reference ROI (R1, R2) received from the ROI rotation angle detection unit 43c. Thus, the rotation angle of each distance image data (S1, S2) and slice image data (S) received from the vertical linear interpolation unit 43e is rotated so as to be restored to the rotation angle before the rotational movement by the ROI rotation angle matching unit 43d. Move. Then, the distance image data (S1, S2) and the slice image data (S) after the rotation angle return are supplied to the ROI barycentric position return unit 43f. As a result, even when a twisted object such as a blood vessel is extracted, it is possible to extract a three-dimensional region (D) in which connectivity is maintained.

  Then, similarly to the case where the rotational movement of each distance image data and slice image data is not performed, in step S14, the center of gravity position of each distance image data and slice image data is converted to the center of gravity before the parallel movement by the ROI center of gravity position returning unit 43f. Translated back to position.

  Next, in step S18, the ROI generated with the parallel movement and the rotational movement in this way is ROI together with each distance image data and slice image data after the return of the center of gravity position as ROI information by the ROI center of gravity position return unit 43f. It is written in the storage unit 44. That is, an ROI for extracting an object such as a blood vessel is set in each slice image data.

  When the ROI is set in this way, in step S7 in FIG. 4, the image extraction unit 45 stores each slice image data stored in the image storage unit 36 based on each ROI stored in the ROI storage unit 44. Of these, only the ROI part is extracted. Then, each slice image data of the extracted ROI portion is written and stored in the extracted image storage unit 46 by the image extraction unit 45.

  However, in addition to extraction processing of slice image data included in the ROI portion, processing such as deletion processing and paint processing may be performed.

  FIG. 10 is a diagram illustrating processing on slice image data by the image extraction unit 45 of the ultrasonic diagnostic apparatus 30 illustrated in FIG. 11.

  For example, the slice image data of the region A as shown in FIG. 10A is read from the image storage unit 36 to the image extraction unit 45, while the ROI of the region B as shown in FIG. 10B is read from the ROI storage unit 44. It is read into the image extraction unit 45. Then, the image extraction unit 45 creates slice image data included in the common part of the regions A and B as shown in FIG. 10C, for example, by performing extraction processing of slice image data included in the ROI part. To do. Further, the image extraction unit 45 is included in the common part between the region A and the region other than the region B as shown in FIG. Slice image data can also be created. Furthermore, the image extraction unit 45 can also create slice image data indicated by the sum of the region A and the region B as shown in FIG.

  Next, in step S8 of FIG. 4, the 3D display processing unit 47 performs superimposition processing on the ROI slice image data stored in the extracted image storage unit 46 and processing for matching the gradation of the display unit 38. A three-dimensional display process such as the above is performed and given to the display unit 38. As a result, organs such as blood vessels and organs are extracted and displayed on the display unit 38 as three-dimensional images such as SVR images, MIP images, and MPR images.

  That is, the ultrasonic diagnostic apparatus 30 described above is equally distanced along with slice image data between distance image data in which the reference ROI is set so that the center of gravity of the reference ROI adjacent in the slice direction is on a straight line in the slice vertical direction. By connecting the image data in the slice direction and setting the 3D region by linear interpolation, the distance image data and the slice image data are restored to their original positions. An area is extracted as an ROI.

  For this reason, even if the position of the center of gravity in the cross section in the slice direction is not on an axis perpendicular to the slice direction, such as an organ such as a blood vessel or an organ, an appropriate ROI is set and a three-dimensional region is set. Can be extracted as

  Furthermore, in addition to the above, the ultrasonic diagnostic apparatus 30 extracts the organs having a twisted relationship between the distance image data in which the reference ROIs are set so that the rotation angles of the reference ROIs adjacent in the slice direction coincide with each other. Rotate and move the distance image data together with the slice image data on a plane parallel to the slice direction, set a three-dimensional region by linear interpolation, and then restore the distance image data and slice image data to their original positions As a result, a three-dimensional region in which connectivity is maintained is extracted as an ROI.

  In other words, according to the ROI interpolation area extracting apparatus as described above, even if the organ exists obliquely with respect to the setting surface of the ROI or the organ has a twisted relationship, it can appropriately translate and rotate. A three-dimensional region can be extracted by linear interpolation.

  The image processing system 35 having such an image processing function is not limited to the ultrasound diagnostic apparatus 30 but is incorporated in an image diagnostic apparatus such as an X-ray CT apparatus or an MRI apparatus, and is a two-dimensional (2D) slice image. It can be applied to ROI setting for reconstructing 3D image data from data.

  In the image processing system 35, the example in which the ROI is set in the slice image data has been shown. However, if the object to be set in the ROI is a plurality of parallel planes, the two-dimensional object of the subject in an arbitrary cross section such as MPR image data is used. An ROI may be set for tomographic image data.

  In addition, a value obtained by adding 1 each time the distance image advances one pixel inside the outline of the reference ROI is used as the value of the pixel, and a value obtained by subtracting 1 every time one pixel moves outside the outline of the reference ROI Although it was created by setting the value of the pixel, a distance image may be created by using a value obtained by adding or subtracting an arbitrary number every time an arbitrary number of pixels advance. Furthermore, when setting the ROI based on the CT image collected by the X-ray CT apparatus, the CT value may be used instead of the pixel value.

  Further, the image processing system 35 may be provided with only the function of the rotational movement process without providing the function of the parallel movement process of the distance image data and the slice image data.

1 is a functional block diagram showing an embodiment of an ultrasonic diagnostic apparatus as an example of an image diagnostic apparatus according to the present invention. The functional block diagram which shows the detailed structure of the ROI interpolation part of the image processing system incorporated in the ultrasonic diagnosing device shown in FIG. The figure explaining the example of the input data of the image processing system incorporated in the ultrasonic diagnosing device shown in FIG. The flowchart which shows the procedure at the time of displaying the image contained in a three-dimensional area | region by the ultrasound diagnosing device shown in FIG. The flowchart which shows the detail of the procedure which sets ROI of each slice image data by the ROI interpolation part in the flowchart shown in FIG. The figure explaining the setting method of ROI of each slice image data by the ROI interpolation part shown in FIG. The figure explaining the method to set ROI by linear interpolation using two distance images. The figure which represented the distance image and interpolation image shown in FIG. The figure explaining the setting method of ROI in case the reference ROI and each slice image data are rotationally moved by the ROI interpolation part shown in FIG. The figure which shows the process with respect to the slice image data by the image extraction part of the ultrasonic diagnosing device shown in FIG. The functional block diagram of the conventional image processing system. The figure explaining the extraction method of the three-dimensional area | region by the ROI interpolation part of the conventional image processing system shown in FIG. The figure explaining an example of the problem which arises when the extraction method of the three-dimensional area | region by the ROI interpolation part of the conventional image processing system shown in FIG. 11 is used. The figure explaining another example of the problem which arises when the extraction method of the three-dimensional area | region by the ROI interpolation part of the conventional image processing system shown in FIG. 11 is used.

Explanation of symbols

DESCRIPTION OF SYMBOLS 30 Ultrasonic diagnostic apparatus 31 Transmission circuit 32 Reception circuit 33 Ultrasonic probe 34 Image generation circuit 35 Image processing system 36 Image storage part 37 Display processing part 38 Display part 39 Coordinate conversion part 40 Mouse 41 Distance image calculation part 42 Distance image storage part 43 ROI interpolation unit 43a ROI centroid position detection unit 43b ROI centroid position matching unit 43c ROI rotation angle detection unit 43d ROI rotation angle matching unit 43e Vertical linear interpolation unit 43f ROI centroid position return unit 43g ROI rotation angle return unit 44 ROI storage unit 45 Image extraction unit 46 Extracted image storage unit 47 3D display processing unit

Claims (8)

Image data collection means for collecting two-dimensional tomographic image data of a subject;
ROI center-of-gravity position detecting means for detecting a position of the center of gravity of the reference ROI with a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data as a reference ROI;
Based on the position of the center of gravity detected by the ROI center of gravity position detecting means, the position of the center of gravity of the reference ROI adjacent to the slice direction of the two-dimensional tomographic image data is on a straight line in the direction perpendicular to the slice. ROI centroid position matching means for equally translating the two-dimensional tomographic image data for which the reference ROI is not set together with the two-dimensional tomographic image data for which the reference ROI is set, in a direction parallel to the slice;
A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the parallel movement by the ROI centroid position matching means, and the three-dimensional region included in the set three-dimensional region Vertical linear interpolation means for setting a portion of the two-dimensional tomographic image data after translation as a region of interest of the two-dimensional tomographic image data;
ROI centroid position return for translating the two-dimensional tomographic image data so that the position of the centroid of the two-dimensional tomographic image data after the parallel movement in which the region of interest is set is restored to the position of the centroid before the parallel movement. Means,
A diagnostic imaging apparatus comprising: Image data collection means for collecting two-dimensional tomographic image data of a subject;
ROI rotation angle detection means for detecting a rotation angle of the reference ROI using a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data as a reference ROI;
Based on the rotation angle detected by the ROI rotation angle detection means, the two-dimensional image in which the reference ROI is set so that the rotation angles of the reference ROIs adjacent in the slice direction of the two-dimensional tomographic image data coincide with each other. ROI rotation angle matching means for rotating the two-dimensional tomographic image data for which the reference ROI is not set together with the tomographic image data evenly on a plane parallel to the slice;
A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the rotational movement by the ROI rotation angle matching means, and the three-dimensional region included in the set three-dimensional region Vertical linear interpolation means for setting a portion of the two-dimensional tomographic image data after rotational movement as a region of interest of the two-dimensional tomographic image data;
ROI rotation angle return means for rotating the two-dimensional tomographic image data so that the rotation angle of the two-dimensional tomographic image data after the rotational movement in which the region of interest is set is returned to the rotation angle before the rotational movement. ,
A diagnostic imaging apparatus comprising: ROI barycentric position detecting means for detecting the position of the barycenter of the reference ROI with a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data of the subject as a reference ROI;
Based on the position of the center of gravity detected by the ROI center of gravity position detecting means, the position of the center of gravity of the reference ROI adjacent to the slice direction of the two-dimensional tomographic image data is on a straight line in the direction perpendicular to the slice. ROI centroid position matching means for equally translating the two-dimensional tomographic image data for which the reference ROI is not set together with the two-dimensional tomographic image data for which the reference ROI is set, in a direction parallel to the slice;
A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the parallel movement by the ROI barycentric position matching means, and the three-dimensional region included in the set three-dimensional region Vertical linear interpolation means for setting a portion of the two-dimensional tomographic image data after translation as a region of interest of the two-dimensional tomographic image data;
ROI centroid position return for translating the two-dimensional tomographic image data so that the position of the centroid of the two-dimensional tomographic image data after the parallel movement in which the region of interest is set is restored to the position of the centroid before the parallel movement. Means,
An image processing system comprising: ROI rotation angle detection means for detecting a rotation angle of the reference ROI using a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data of the subject as a reference ROI;
Based on the rotation angle detected by the ROI rotation angle detection means, the two-dimensional image in which the reference ROI is set so that the rotation angles of the reference ROIs adjacent in the slice direction of the two-dimensional tomographic image data coincide with each other. ROI rotation angle matching means for rotating the two-dimensional tomographic image data for which the reference ROI is not set together with the tomographic image data evenly on a plane parallel to the slice;
A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the rotational movement by the ROI rotation angle matching means, and the three-dimensional region included in the set three-dimensional region Vertical linear interpolation means for setting a portion of the two-dimensional tomographic image data after rotational movement as a region of interest of the two-dimensional tomographic image data;
ROI rotation angle return means for rotating the two-dimensional tomographic image data so that the rotation angle of the two-dimensional tomographic image data after the rotational movement in which the region of interest is set is returned to the rotation angle before the rotational movement. ,
An image processing system comprising: Computer
ROI barycentric position detecting means for detecting the position of the barycenter of the reference ROI using a plurality of regions of interest serving as the reference set in the two-dimensional tomographic image data of the subject as a reference ROI;
Based on the position of the center of gravity detected by the ROI center of gravity position detecting means, the position of the center of gravity of the reference ROI adjacent to the slice direction of the two-dimensional tomographic image data is on a straight line in the direction perpendicular to the slice. ROI centroid position matching means for equally translating the two-dimensional tomographic image data for which the reference ROI is not set together with the two-dimensional tomographic image data for which the reference ROI is set, in a direction parallel to the slice;
A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the parallel movement by the ROI barycentric position matching means, and the three-dimensional region included in the set three-dimensional region Vertical linear interpolation means for setting a portion of the two-dimensional tomographic image data after translation as a region of interest of the two-dimensional tomographic image data;
And the ROI centroid position for translating the two-dimensional tomographic image data so that the position of the centroid of the two-dimensional tomographic image data after the parallel movement in which the region of interest is set is returned to the position of the centroid before the parallel movement. Return means,
An image processing program that functions as an image processing program. Computer
ROI rotation angle detection means for detecting a rotation angle of the reference ROI using a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data of the subject as a reference ROI;
Based on the rotation angle detected by the ROI rotation angle detection means, the two-dimensional image in which the reference ROI is set so that the rotation angles of the reference ROIs adjacent in the slice direction of the two-dimensional tomographic image data coincide with each other. ROI rotation angle matching means for uniformly rotating and moving the two-dimensional tomographic image data in which the reference ROI is not set together with the tomographic image data on a plane parallel to the slice;
A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the rotational movement by the ROI rotation angle matching means, and the three-dimensional region included in the set three-dimensional region Vertical linear interpolation means for setting a portion of the two-dimensional tomographic image data after rotational movement as a region of interest of the two-dimensional tomographic image data;
And ROI rotation angle return means for rotating the two-dimensional tomographic image data so that the rotation angle of the two-dimensional tomographic image data after the rotational movement in which the region of interest is set is returned to the rotation angle before the rotational movement. ,
An image processing program that functions as an image processing program. Detecting a position of the center of gravity of the reference ROI with a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data of the subject as the reference ROI;
Based on the detected position of the center of gravity, the reference ROI is set so that the position of the center of gravity of the reference ROI adjacent in the slice direction of the two-dimensional tomographic image data is on a straight line in the direction perpendicular to the slice. Parallelly moving the two-dimensional tomographic image data in which the reference ROI is not set together with the two-dimensional tomographic image data performed in a direction parallel to the slice;
A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the parallel movement, and the two after the parallel movement included in the set three-dimensional region. Setting a portion of the two-dimensional tomographic image data as a region of interest of the two-dimensional tomographic image data;
Translating the two-dimensional tomographic image data so that the position of the center of gravity of the two-dimensional tomographic image data after the parallel movement in which the region of interest is set is returned to the position of the center of gravity before the parallel movement;
An image processing method comprising: Detecting a rotation angle of the reference ROI with a plurality of regions of interest serving as a reference set in the two-dimensional tomographic image data of the subject as the reference ROI;
Based on the detected rotation angle, the reference ROI is set together with the two-dimensional tomographic image data in which the reference ROI is set such that the rotation angles of the reference ROIs adjacent in the slice direction of the two-dimensional tomographic image data match. Rotating the two-dimensional tomographic image data for which is not set evenly on a plane parallel to the slice;
A three-dimensional region is set by performing linear interpolation in a direction perpendicular to the slice of the two-dimensional tomographic image data after the rotational movement, and the two after the rotational movement included in the set three-dimensional region. Setting a portion of the two-dimensional tomographic image data as a region of interest of the two-dimensional tomographic image data;
Rotationally moving the two-dimensional tomographic image data so that the rotational angle of the two-dimensional tomographic image data after the rotational movement in which the region of interest is set is restored to the rotational angle before the rotational movement;
An image processing method comprising:

Images