JP2007111123A - Medical image display device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To highly precisely extract an intended target internal organ area from a volume image of a subject. <P>SOLUTION: This medical image display device is provided with an automatic mode automatically finding an image number (a slice number) where the target internal organ area is magnifiedly imaged, a large area mode where an operator selects the image number, a control line mode selecting a target internal organ area included in a plurality of tomograms and setting a control line formed by connecting them, and a trace mode tracing a profile line of the target internal organ area with a mouse and automatically extracting the inside of the traced profile line as the target internal organ area. The target internal organ extracted in the respective modes is two-dimensionally or three-dimensionally displayed. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a medical image display apparatus, and more particularly to a technique for extracting a desired organ region from a tomographic image of a subject.

  When organ transplantation such as the liver is performed, the donor (donor) liver is cut according to the liver volume of the patient to be transplanted, and it is necessary to know the volume of the liver. Therefore, it is necessary to take a volume image of each part of the patient and donor including the liver and extract the liver region from the volume image.

  However, since the liver is in contact with other organs such as the kidney, it is conventionally known that organs other than the target organ can be extracted only by simple threshold processing.

As a technique for solving this problem, for example, Patent Document 1 proposes a method of performing volumetric imaging of a patient's liver region twice and automatically extracting a liver region using image data of two time phases. .
Japanese Patent Laid-Open No. 2002-345807

  However, the extraction method of Patent Document 1 has a problem that the patient is photographed twice, and the amount of exposure to the patient increases.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a medical image display device capable of accurately extracting only a target organ region from a volume image obtained by one imaging. .

  In order to solve the above problem, the medical image display apparatus according to the present invention starts extraction processing of the target organ region from a volume image obtained by stacking a plurality of tomographic images obtained by imaging the target organ of the subject. Identifying means for identifying a starting tomographic image that is a tomographic image, first extracting means for extracting the target organ region included in the starting tomographic image, and recording means for recording the extracted target organ region as a reference image Reading means for reading other tomographic images adjacent to the starting tomographic image; second extracting means for extracting the target organ region from the read other tomographic images based on the correlation with the reference image; Display means for displaying the target organ area extracted by the first extraction means and the second extraction means, and the recording means uses the target organ area extracted by the second extraction means as a new reference image. New recording is performed, the reading unit reads a new other tomographic image adjacent to the other image, and the second extracting unit reads the new reference tomographic image from the new other tomographic image based on the updated reference image. The target organ region is extracted.

  Here, “based on the correlation with the reference image” includes the correlation with the shape of the reference image and the continuity (coordinate correlation) between the tomographic image including the reference image and other tomographic images.

  Further, “displaying the target organ region” includes a case of displaying as a two-dimensional image and a case of displaying as a pseudo three-dimensional image.

  The specifying unit calculates, for each of the plurality of tomographic images, the number of pixels satisfying the density value of the target organ included in each tomographic image, and specifies the tomographic image having the maximum number of pixels as a starting tomographic image. May be.

  The display unit displays the plurality of tomographic images, and the specifying unit selects a tomographic image for starting extraction processing of the target organ region from the displayed plurality of tomographic images; Designating means for designating an arbitrary point included in the target organ area in the selected tomographic image, and extracting an area including the specified arbitrary point, and the selected area is selected based on the extracted area And validity determination means for determining whether or not the tomographic image is valid as a starting tomographic image for starting the extraction processing of the target organ region.

  Further, the target organ region is a liver region in which the liver of the subject is imaged, and the validity determination unit includes a spine region extraction unit that extracts a spine region included in the selected tomographic image, Coordinate calculating means for calculating the barycentric coordinates of the extracted spine region and the maximum and minimum values of the horizontal coordinate of the region including the arbitrary point extracted from the selected tomographic image; Range determining means for determining whether or not the horizontal axis coordinate of the barycentric coordinate of the region belongs to a range larger than the minimum value of the coordinate in the horizontal axis direction of the region including the arbitrary point and smaller than the maximum value; Warning means may be provided that issues a warning when the determination means determines that the horizontal axis coordinates of the barycentric coordinates of the spinal region do not belong to the range.

  Further, the validity determination means determines the validity based on a comparison result between the density value of the region including the arbitrary point extracted from the selected tomographic image and the density value of the target organ. May be.

  The display unit displays the plurality of tomographic images, and the specifying unit selects a tomographic image for starting extraction processing of the target organ region from the displayed plurality of tomographic images; Contour line input means for inputting the contour line of the target organ region in the selected tomographic image, and the density value of the other region other than the target organ region in the selected tomographic image Locking means for changing the density value to a density value different from the density value, and the first extracting means extracts an inner area of the contour line after the density value of the other area is changed by the locking means. May be.

  The display unit displays the plurality of tomographic images, and the specifying unit selects a tomographic image for starting extraction processing of the target organ region from the displayed plurality of tomographic images; Correction means for correcting the outline of the internal area extracted by the first extraction means, the display means displays the extracted internal area, and the correction means is arranged on the displayed internal area. The internal region of the contour line corrected in step may be re-extracted as the target organ region.

  The medical image display apparatus according to the present invention includes a display unit that displays the tomographic image constituting a volume image obtained by stacking a plurality of tomographic images obtained by imaging a target organ of a subject, and the displayed Selection means for selecting a plurality of starting tomographic images for starting extraction processing of the target organ region from a plurality of tomographic images, and a determination point for a point included in the target organ region on each of the selected starting tomographic images In the volume image, a line connecting the determination points set on the plurality of starting tomographic images is temporarily set, and an angle formed by the temporarily set line and a plane including the tomographic image Determination line setting means for setting a line within a predetermined angle range for passing through the inside of the target organ region included in the volume image as a determination line, and a density value of the target organ for each tomographic image Based on candidate region extraction means for extracting at least one region including the target organ region as a candidate region, and the extracted candidate region that intersects with the determination line as the target organ region And a target organ region extracting means for extracting, wherein the display means further displays the extracted target organ region.

  Further, the image processing apparatus further includes a template indicating the shape of the target organ region, the display means displays the template and the starting tomographic image selected by the selection means so as to overlap each other, and the input means uses the template as a reference. The determination point may be input by setting the relative coordinates.

  According to the present invention, a target organ region is extracted from tomographic images constituting a volume image. Next, another tomographic image adjacent to the tomographic image from which the target organ region is extracted is read, and the target organ region is sequentially extracted from the adjacent tomographic image. Therefore, a desired target organ region can be extracted from the volume image obtained by one imaging. In particular, when a volume image is taken using an X-ray CT apparatus, it is possible to prevent an increase in the amount of exposure due to the second taking.

  Hereinafter, preferred embodiments of a medical image display apparatus according to the present invention will be described in detail with reference to the accompanying drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment of the invention, and the repetitive description thereof is omitted. In the following embodiments, the liver of a subject is imaged by an X-ray CT apparatus, and a plurality of tomographic images (hereinafter referred to as “volume images”) along the body axis direction are obtained. Each tomographic image constituting this volume image is called a slice, and an image number (slice number) is assigned to each slice along the body axis direction. A liver region is extracted from the volume image and displayed. The medical imaging apparatus is not limited to the X-ray CT apparatus, and may be any apparatus that can capture a tomographic image of a subject such as an MRI apparatus.

<Hardware configuration>
FIG. 1 is a hardware configuration diagram showing a configuration of a medical image display system 1 according to an embodiment of the invention. The medical image display system 1 in FIG. 1 includes an X-ray CT apparatus 2, an MR apparatus 3, an image database 4 that stores medical images obtained by the X-ray CT apparatus 2 and the MR apparatus 3, and medical images. The X-ray CT apparatus 2, the MR apparatus 3, the image database 4, and the medical image display apparatus 10 are connected to each other via a network such as a LAN 5.

  The medical image display apparatus 10 includes a central processing unit (CPU) 11 that mainly controls the operation of each component, a main memory 12 that stores a control program for the apparatus and that serves as a work area when the program is executed, and an operating system. System (OS), peripheral device drive, magnetic disk 13 for storing various application software including a program for extracting a target organ region from medical images, and display data are temporarily stored. A display memory 14, a monitor 15 such as a CRT monitor or a liquid crystal monitor for displaying an image based on data from the display memory 14, a keyboard 16, a mouse 17 as a position coordinate input device, and a state of the mouse 17 are detected. Then, signals such as the position of the mouse pointer on the monitor 15 and the state of the mouse 17 are sent to the CPU 11. And an output to the controller 18, and a common bus 19 for connecting the above components.

  In the present embodiment, the medical image display device 10 reads a medical image from the image database 4 via the LAN 5, but a storage device connected to the medical image display device 10, such as an FDD, CD-RW drive, or MO drive A medical image may be read from a ZIP drive or the like. Alternatively, a medical image may be acquired directly from the X-ray CT apparatus 2 or the MR apparatus 3 via the LAN 5.

  FIG. 2 is a program block diagram for performing processing according to the present embodiment. The program includes a mode selection unit 11a, a tomographic image specification unit 11b, a reference image extraction unit 11c, a reference image recording unit 11d, a reading unit 11e, a shape extraction unit 11f, a display control unit 11g, a starting tomographic image selection unit 11h, and validity determination. A unit 11i, a continuity extraction unit 11k, a determination line setting unit 11l, a candidate region extraction unit 11m, a target organ region extraction unit 11n, a template processing unit 11o, a protection unit 11p, an internal region extraction unit 11q, and a correction unit 11r. .

  The mode selection unit 11a is a program for receiving selection of four extraction modes for extracting a liver region, “automatic mode”, “large region mode”, “decision line mode”, and “trace mode”.

  The tomographic image specifying unit 11b is a program used in the “automatic mode”, and automatically sets a starting tomographic image.

  The reference image extraction unit 11c, the reference image recording unit 11d, the reading unit 11e, the shape extraction unit 11f, and the continuity extraction unit 11k use the liver region extracted from the starting tomographic image as a reference image, and sequentially extract liver regions from other images. It is a program to do. The shape extraction unit 11f performs extraction processing based on the correlation value with the shape of the reference image. The continuity extraction unit 11k extracts the liver region using the continuity between the starting tomographic image including the reference image and other tomographic images, that is, the correlation of coordinates. The shape extraction unit 11f and the continuity extraction unit 11k can be selectively used.

  The start tomogram selection unit 11h and the validity determination unit 11i are programs mainly used in the “large area mode”, and manually set the start tomogram.

  The decision line setting unit 11l, the candidate region extraction unit 11m, the target organ region extraction unit 11n, and the template processing unit 11o are programs mainly used in the “determination line mode”. This is a program to extract.

  The protect unit 11p, the internal region extraction unit 11q, the correction unit 11r, and the threshold value extraction unit 11s are programs mainly used in the “trace mode”, and the outline of the liver region is displayed in the start tomographic image selected by the operator. This is a program for tracing and automatically extracting a liver region based on the trace information, and correcting the automatically extracted liver region based on the trace information.

  These programs are stored in the HDD 13, loaded into the memory 12 as appropriate, and executed by the CPU 11.

<Outline of processing>
A top screen 30 shown in FIG. 3 is displayed on the monitor 15 of the medical image display apparatus 10. On the top screen 30, an automatic mode selection button 31, a large area mode selection button 32, a determination line mode selection button 33, and a trace mode selection button 34 that are icons for selecting four extraction modes are displayed. When the operator clicks any one of the mode selection buttons 31 to 34 with the mouse 17, the processing of the icon starts. On the top screen 30, a tomographic image 35 constituting a volume image is displayed.

  FIG. 4 is a flowchart showing an operation procedure of liver extraction processing in the medical image display apparatus 10.

  In S1, the mode selection unit 11a determines whether or not the automatic mode is selected (S1). If the automatic mode is selected, the process proceeds to S2. If not selected, the process proceeds to S3.

  In S2, extraction processing in the automatic mode is performed. In the automatic mode, a large area detection process for automatically obtaining an image number (slice number) in which a large liver area is captured and a liver area extraction process using continuity with other slices are executed ( S2).

  In S3, the mode selection unit 11a determines whether or not the large area mode is selected (S3). If the large area mode is selected, the process proceeds to S4, and if not selected, the process proceeds to S5.

  In S4, extraction processing in the large area mode is performed. In the large area mode, the operator selects a slice in which the liver area is photographed in a large size, and manually inputs the coordinates of the liver area included in the slice using the mouse 17 and other slices. A liver region extraction process using continuity is executed (S4).

  In S5, the mode selection unit 11a determines whether or not the determination line mode is selected (S5). If the decision line mode is selected, the process proceeds to S6, and if it is not selected, the process proceeds to S7.

  In S6, extraction processing in the determination line mode is performed. In the decision line mode, at least two slices are selected, a decision point setting process for setting decision points in the liver area included in each slice, and a decision line connecting these decision points are created, and the region including this decision line Is extracted as a liver region. (S6).

  In S7, the mode selection unit 11a determines whether or not the trace mode is selected (S7). When the trace mode is selected, the process proceeds to S8, and when it is not selected, the process is terminated.

  In S8, extraction processing in the trace mode is performed. Trace mode is the process of tracing the outline of the liver area with the mouse and automatically extracting the traced outline as the liver area. Before this automatic extraction, the outline is set inside the liver area. In order to prevent this, an extraction protection mode that prohibits selection based on the CT value of the liver region and a correction mode that corrects excessive or insufficient extraction after automatic extraction are included. These modes are selected by clicking from the list after clicking a trace mode selection button 34 in FIG. 3 to be described later.

<Auto mode>
FIG. 5 is a flowchart showing a processing procedure in the automatic mode. This process starts when the mode selection unit 11a determines that the automatic mode is selected.

  In step S201, the tomographic image specifying unit 11b obtains a slice number of an image in which the number of pixels having a CT value corresponding to the liver region is maximized from the volume image (S201).

The minimum value (T1) and the maximum value (T2) of the CT value corresponding to the liver region can be determined from substances constituting the liver region. Therefore, the tomographic image specifying unit 11b obtains the number of pixels satisfying Equation 1 for each slice.
[Equation 1]
T1 <CT value <T2
Next, the tomographic image specifying unit 11b obtains a slice number (referred to as “maximum slice number”) that maximizes the number of pixels. Thereby, it is possible to automatically specify the tomographic image in which the liver region is photographed the largest from the volume image as the starting tomographic image.

  In step S202, the reference image extraction unit 11c extracts a liver region from the start tomogram (tomogram with the maximum slice number) obtained in S201 (S202).

  The reference image extraction unit 11c performs threshold processing based on the CT value of the liver for the starting tomographic image to generate a binary image.

  Immediately after simple threshold processing, a liver region and an organ region adjacent thereto, for example, a kidney region, are connected and extracted by a thin curved region. Therefore, the reference image extraction unit 11c performs a process of deleting a partial protruding region of the binary image in order to cut the liver region from other organ regions. This deletion process is shown in FIG. In FIG. 6, point sequences a1, a2, a3, a4, and a5 that are parallel to the x axis and that are composed of a plurality of pixel points are set, and if a1 = a5 = 0, a2 = a3 = a4 = 0. In addition, point sequences b1, b2, b3, and b4 made up of a plurality of pixel points parallel to the y-axis are set. If b1 = b4 = 0, b2 = b3 = 0. As a result, the point sequence a1, a2, a3, a4, a5 and the point sequence b1, b2, b3, b4 are cut or deleted, and the partial protruding area of the binary image is deleted. As a result, it is possible to cut organ regions extracted by being connected by thin curved regions. By this processing, a plurality of cut organ regions are extracted as shown in FIG. The plurality of organ regions include other organ regions such as a kidney region in addition to a liver region. The above cutting process is an example, and the present invention is not limited to this.

  The reference image extraction unit 11c performs processing for specifying a liver region from these organ regions. The liver is a large organ among human organs, and is located on the left in the abdominal axial image. The liver region is specified using the characteristics of the liver.

  The following conditions are used for specifying the liver region.

  Condition 1: The number of pixels constituting each organ region extracted and separated in FIG. 6 is measured, and the organ region having the maximum number of pixels is extracted.

  Condition 2: The X-coordinate centroid of each organ region extracted and separated in FIG. 6 is calculated. Then, the organ region having the largest (image size-X coordinate centroid), that is, the organ region where the centroid is located most to the left in the image of FIG.

  Those satisfying condition 1 and condition 2 may be extracted as a liver region, and if one satisfying either condition 1 or condition 2 is determined, this may be extracted as a liver region.

  In step S203, the reference image recording unit 11d records the liver region extracted in S202 in the extracted image storage region in the memory 12, generates a copy thereof, and records the copy in the reference image storage region in the memory 12 as a reference image. (S203).

  In step S204, the reading unit 11e reads a tomographic image having a slice number that is one greater than the slice number in S201 (S204). As a result, the tomographic image adjacent to the starting tomographic image and having a slice number that is one greater than the starting tomographic image is read.

  In step S205, a liver region is extracted based on the correlation with the reference image (S205). This extraction process may use the correlation of the shape of the reference image, or may use the continuity between slices. Here, the shape extraction unit 11f extracts the liver region based on the correlation value between the shape of the reference image and the shape of the organ region included in the tomographic image read in S204.

  In step S206, the reading unit 11e determines whether the image having the largest slice number has been read. If “Y”, the process proceeds to S209, and if “N”, the process proceeds to S207.

  In step S207, the reference image recording unit 11d records the liver area extracted in S205 in the extracted image storage area of the memory 12. Further, a copy of the liver area is generated, and the reference image stored in the reference image storage area of the memory 12 is overwritten (updated) in S203.

  In step S208, the reading unit 11e reads a slice having a number larger than the slice number in S205 (S208). Then, the process returns to S205, and the liver region is extracted from the tomographic image read in S208 based on the updated reference image.

  In step S209, a tomographic image having a slice number that is the smallest than the slice number in S201 is read (S209). As a result, a tomographic image adjacent to the starting tomographic image and having a slice number that is one smaller than the starting tomographic image is read.

  In step S210, the reference image recording unit 11d reads out the liver area extracted from the start tomogram from the memory 12, generates a copy, and overwrites the reference image storage area of the memory 12 with the copy.

  In step S211, the shape extraction unit 11f extracts the liver region from the image read in S209 based on the correlation with the shape of the reference image in S210, as in S205. The extracted image is recorded in the extracted image storage area of the memory 12.

  In step S212, the reading unit 11e determines whether an image having the smallest slice number has been read. If “Y”, the process proceeds to S215, and if “N”, the process proceeds to S213.

  In step S213, the reference image recording unit 11d overwrites (updates) the reference image in the same manner as in S207.

  In step S214, the reading unit 11e reads a slice having a number smaller than the slice number in S211 (S214). Then, the process returns to S211, and the liver region is extracted from the tomographic image read in S214 based on the updated reference image.

  In step S215, the display control unit 11g displays the extracted images sequentially stored in the extracted image storage area of the memory 12 on the monitor 15 as a liver area (S215). The display control unit 11g may generate and display a pseudo three-dimensional image obtained by stacking liver regions extracted from each stored slice, or display the liver region extracted from each slice as a two-dimensional image. May be.

  In the above, the liver extraction processing is first performed in the direction in which the image number increases with the starting tomogram as the center, and then the liver region extraction processing is performed in the direction in which the image number decreases, but this order is reversed. May be. Further, the direction in which the image number increases and the direction in which the image number decreases may be performed simultaneously with the starting tomographic image as the center. Further, the direction in which the image number is increased and the direction in which the image number is decreased may be alternately advanced around the starting tomographic image.

  According to this automatic mode, a starting tomographic image in which the largest liver region is imaged is identified from the volume image obtained by stacking a plurality of tomographic images, and the liver region is extracted from the identified starting tomographic image. Then, using this liver region as a reference image, a liver region of an adjacent tomographic image is extracted. Using the liver region extracted from the adjacent tomographic image as a new reference image, the process of extracting the liver region from the adjacent tomographic image is repeated. Thereby, a liver region can be automatically extracted from a volume image obtained by stacking a plurality of tomographic images.

<Large area mode>
FIG. 7 is a flowchart showing a processing procedure in the large area mode. This process starts when the mode selection unit 11a determines that the large area mode has been selected.

  In step S401, the tomographic image selection unit 11h displays a plurality of slices constituting the volume image on the monitor 15. The operator clicks with the mouse 17 to select a tomographic image showing the largest liver region from among the displayed tomographic images.

  In step S402, the coordinates of points included in the liver region in the tomographic image selected in step S401 by the operator are input by the mouse 17. The keyboard 16 may be used for coordinate input.

  In step S403, the validity determination unit 11i extracts a region including the coordinates input in S402.

  In step S404, the validity determination unit 11i determines whether the tomographic image selected in S401 is valid as a starting tomographic image based on the region extracted in S403. The process of the validity judgment part 11i is demonstrated based on FIG.

  In step S4041, the validity determination unit 11i determines whether the region extracted in S404 is a liver region. This determination is based on whether the pixel value of the extracted region satisfies the pixel value corresponding to the CT value of the liver.

  In step S4042, the validity determination unit 11i obtains the minimum value (Xmin) and maximum value (Xmax) in the horizontal axis direction of the region extracted in S404 (S4041).

  In step S4043, the validity determination unit 11i extracts a spinal region from the tomographic image selected in S401. Extraction of the spine region is performed using threshold processing based on the CT value of the spine. The validity determination unit 11i obtains the barycentric coordinates (Xo, Yo) of the spinal region.

In step S405, the x coordinate of the centroid coordinates of the spine region obtained by the validity determination unit 11i in S4042 satisfies the formula 2, that is, from the minimum value (Xmin) to the maximum value (Xmax) obtained in S4041. Judge whether it belongs to the range.
[Formula 2]
Xmin <Xo <Xmax
If the validity determination unit 11i determines “belongs”, the process proceeds to S406 as “Y”. If it is determined that it does not belong, the process proceeds to S407 as “N”.

  In step S406, the continuity extraction unit 11k extracts a liver region from another tomographic image using continuity between slices. Based on FIG. 9, an extraction process using continuity of slices will be described.

  The volume image is composed of a plurality of tomographic images with slice numbers n-1, n, n + 1,. The tomographic image selected in S402 is set as the tomographic image 60 of slice number n. It is assumed that the region 70 included in the tomographic image 60 is determined to be valid as a liver region in S404 and S405, and the tomographic image 60 is determined to be valid as a starting tomographic image. The reading unit 11e reads a tomographic image 61 (slice number n + 1) adjacent to the slice number n.

  The continuity extracting unit 11k performs threshold processing corresponding to the density value of the liver on the tomographic image 61, and extracts a plurality of organ regions.

  Next, the continuity extracting unit 11k has a high correlation value between the coordinates of the region 70 included in the tomographic image 60 among the plurality of organ regions 71a and 71b included in the tomographic image 61, that is, there is continuity with the region 70. The region 71a is extracted as a liver region.

  In step S407, the validity determination unit 11i displays a warning to select the tomographic image again, and the process returns to S401.

  In step S408, the display control unit 11g displays the extracted liver region on the monitor 15.

  According to the large area mode, the operator selects an area to start the extraction process of the liver area, checks whether the selected tomographic image is appropriate as the starting tomographic image, and then starts the extraction process of the liver area. can do.

  In step S406, the liver region is extracted from the other tomographic image using the continuity of coordinates between the tomographic images. However, the other tomographic image is updated by the method of updating the reference image described in the automatic mode. You may perform the extraction process from.

  In the above description, the validity determination unit 11i determines the validity based on the barycentric coordinates of the spine region. However, the concentration value of the extracted region is compared with the concentration value of the extracted region and the concentration value of the liver region. May be determined to be appropriate if it is included in the concentration value range of the liver region.

<Judgment line mode>
FIG. 10 is a flowchart illustrating a processing procedure in the determination line mode. This process starts when the mode selection unit 11a determines that the determination line mode has been selected.

  In step S601, the tomographic image selection unit 11h displays a plurality of slices constituting the volume image on the monitor 15. The operator selects a plurality of tomographic images by clicking with the mouse 17 among the displayed tomographic images. As shown in FIG. 11, the plurality of tomographic images selected here are a tomographic image S1 (FIG. 11 (a)) near the upper end of the liver, a tomographic image near the center of the liver, and a lower end of the liver. It is desirable to include the tomographic image Sn (FIG. 11B).

  In step S602, the coordinates of points included in the liver region in each tomographic image selected by the operator in S401 are input by the mouse 17. A keyboard may be used for coordinate input. FIG. 11 is a schematic diagram showing the process of this step. The tomographic image S1 is displayed on the monitor 15, and the mouse 17 is used to click on an arbitrary point in the liver region included in the tomographic image S1 to designate coordinates. Similar processing is performed for the tomographic image Sn.

  In step S603, as in S4041, the validity determination unit 11i determines whether the input point is a liver region. If “Y”, the process returns to S604, and if “N”, the process returns to S601. This step may be omitted and the process may advance from S602 to S604.

  In step S604, the determination line setting unit 11l sets a determination line.

In S602, a line connecting the input points on each tomographic image is temporarily set. Next, it is determined whether or not the angle formed by the temporarily set line and the plane including the tomographic image is within a predetermined angle range for passing through the inside of the target organ region. FIG. 12 is a schematic diagram showing determination lines. Lines 80a, 80b, 80c, 80d, and 80e are provisionally set based on the slice Sn. When the angle α formed by the temporarily set lines 80a, 80b, 80c, 80d, and 80e and the slice Sn satisfies Expression 3, the determination line setting unit 11l sets the temporarily set line as a determination line.
[Equation 3] 0 ≦ α <180
In FIG. 12, the angles formed by the lines 80a, 80b, 80c, and 80d and the slice Sn satisfy Expression 3, so that they are set as determination lines. Since the angle α formed by the temporarily set line 80e and the slice Sn exceeds 180 degrees, it is not set as a determination line. Note that the predetermined angle range expressed by the equation 3 is determined by the relative position of the reference slice in the volume image, and is not limited to the above numerical range.

  In step S605, the candidate area extraction unit 11m performs threshold processing corresponding to the density value of the liver area for each tomographic image, and extracts a candidate area of the liver area for each tomographic image.

  In step S606, the target organ region extraction unit 11n extracts a candidate region that intersects with the determination line from the candidate regions extracted in step S605 as a liver region.

  In step S607, the extracted liver region is displayed.

  According to the determination line mode, a determination line connecting points of liver regions included in a plurality of tomographic images selected by the operator is set, and a region including an intersection of another tomographic image and this determination line is extracted as a liver region. can do.

<Template mode>
The template mode is a mode in which determination points in the determination line mode are set using a template.

  In the template mode, the template processing unit 11o displays the template 100 shown in FIG. A template 100 in FIG. 13 is a template showing the shape of the center of the liver. In the template 100, the upper left corner is defined as coordinates (x0, y0), and the coordinates of determination points to be set are defined as relative coordinates (dx1, dy1), (dx2, dy2).

Next, as in step S601, a tomographic image is displayed on the monitor 15, and a tomographic image in which the liver region is captured in a shape close to the shape of the template 100 is selected. Then, the template 100 is displayed so as to overlap the selected tomographic image. FIG. 14 is a screen for displaying a tomographic image superimposed on the template 100. The template processing unit 11o extracts a region including the CT value of the liver included in the tomographic image displayed superimposed on the template 100. Then, the entire area of the extracted region and the partial area overlapping the template 100 in the extracted region are extracted. If the value obtained by dividing the total area by the partial area is less than a certain value, the process of overlapping the template 100 is performed again. Equation 4 shows this processing.
[Equation 4]
(Area of the region including the CT value of the liver / Area inside the template 100) <Constant value The constant value here is a value indicating the degree of overlap between the template 100 and the tomogram, and the operator is constant. A value may be set.

  FIG. 15A is a screen for displaying the template 101 showing the shape near the upper end of the liver and allowing the operator to select a tomographic image having a high correlation with the template 101. When the operator selects a tomographic image and the template processing unit 11o determines that Expression 4 is satisfied, the template processing unit 11o uses the relative coordinates (FIG. 15B) defined in the template 101, Judgment points are automatically set on the tomographic image.

  FIG. 16 shows a template 102 showing the shape near the lower end of the liver. Also in the case of the template 102 in FIG. 16, determination points are automatically set on the tomographic image by the template processing unit 11o as in FIG.

  By using this template, it is possible to determine whether or not the tomographic image selected by the operator is appropriate as a tomographic image for setting a determination point. In addition, if it is determined that it is appropriate, the determination point can be automatically set.

<Trace mode>
The trace mode starts when the mode selection unit 11a determines that the trace mode has been selected.

  When the operator selects the trace mode, the mode selection unit 11a displays a trace mode selection screen 170 shown in FIG. The trace mode selection screen 170 is provided with a “protect mode” icon 171 and a “correction mode” icon 172. The operator clicks the “protect mode” icon 171 or the correction mode icon 172 to select a mode. A tomographic image 173 is also displayed. A tomographic image 173 is also displayed.

  This will be described below in the order of steps in FIG. FIG. 18 is a flowchart for explaining the flow of processing in the trace mode.

  In step S801, the mode selection unit 11a determines whether or not the protect mode is selected. If the protect mode is selected, the process proceeds to S802, and if it is not selected, the process proceeds to S805.

  In step S802, the operator traces the outline of the liver region with the mouse 17 on the tomographic image.

  In step S803, the protector 11p performs a lock process for preventing an area other than the liver area included in the tomographic image, for example, a kidney area from being extracted in an automatic extraction process performed later. This lock processing is performed by changing the pixel value of the pixels constituting the area other than the liver area to a density value outside the density value range of the liver area.

  In step S804, the internal region extraction unit 11q automatically extracts the internal region of the contour line set in S803 as a liver region.

  In step S805, the threshold extraction unit 11s performs threshold processing based on the CT value of the liver and automatically extracts a liver region from the tomographic image.

  In step S806, the mode selection unit 11a determines whether the correction mode has been selected. If the correction mode is selected, the process proceeds to S807. If the correction mode is not selected, the process proceeds to S808.

  In step S806, the operator traces the outline of the liver region with the mouse 17 on the tomographic image.

  In step S807, the operator traces the contour line of the liver region, and the correcting unit 11r re-extracts the internal region of the traced contour line. Thereby, excess and deficiency of the automatically extracted liver region can be corrected.

  In step S808, the extracted liver region is displayed.

  According to the trace mode, it is possible to automatically extract based on the liver area traced by the operator, or to correct the excess or deficiency of the automatically extracted liver area based on the traced outline.

Hardware configuration diagram showing the configuration of the medical image display system Program block diagram Example of the top screen displayed at the beginning of the liver extraction process The flowchart which shows the operation | movement procedure of a liver extraction process Flow chart showing processing procedure in automatic mode Schematic diagram showing extraction area cutting processing Flow chart showing processing procedure in large area mode The flowchart which shows the process sequence of the validity judgment part 11i. Schematic diagram showing extraction processing using continuity of slices Flowchart showing processing procedure in decision line mode Schematic diagram showing the process of specifying the points included in the liver region Schematic diagram showing judgment lines Schematic diagram showing an example template Schematic showing how the template in the center of the liver is used Schematic showing the usage state of the template at the upper end of the liver Schematic showing the usage state of the template at the lower end of the liver Screen display example of trace mode selection screen 170 Flowchart showing processing procedure in trace mode

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Medical image display system, 2 ... X-ray CT apparatus, 3 ... MR apparatus, 4 ... Image database, 5 ... LAN, 10 ... Medical image display apparatus, 11 ... CPU, 12 ... Main memory, 13 ... Magnetic disk, 14 ... display memory, 15 ... monitor, 16 ... keyboard, 17 ... mouse, 18 ... controller, 19 ... common bus

Claims (9)

A specifying means for identifying a starting tomographic image that is a tomographic image for starting extraction processing of the target organ region out of a volume image obtained by stacking a plurality of tomographic images obtained by imaging a target organ of a subject;
First extraction means for extracting the target organ region included in the starting tomographic image;
Recording means for recording the extracted target organ region as a reference image;
Reading means for reading another tomographic image adjacent to the starting tomographic image;
Second extraction means for extracting the target organ region from the other read tomographic images based on the correlation with the reference image;
Display means for displaying the target organ region extracted by the first extraction means and the second extraction means,
The recording means updates and records the target organ region extracted by the second extracting means as a new reference image, and the reading means reads a new other tomographic image adjacent to the other image, and The extraction means extracts the target organ region based on the updated reference image from the new other tomographic image.
A medical image display device characterized by that. The specifying unit calculates, for each of the plurality of tomographic images, the number of pixels satisfying the density value of the target organ included in each tomographic image, and specifies a tomographic image having the maximum number of pixels as a starting tomographic image;
The medical image display apparatus according to claim 1. The display means displays the plurality of tomographic images;
The specifying means is a selecting means for selecting a tomographic image for starting extraction processing of the target organ region from the displayed tomographic images;
Designating means for designating an arbitrary point included in the target organ region in the selected tomographic image;
Whether an area including the specified arbitrary point is extracted, and based on the extracted area, whether the selected tomographic image is valid as a starting tomographic image for starting the extraction processing of the target organ area And validity judgment means for judging whether or not,
The medical image display apparatus according to claim 1. The target organ region is a liver region in which the liver of the subject is imaged,
The validity determination means includes: a spine area extracting means for extracting a spine area included in the selected tomographic image;
Coordinate calculating means for calculating the barycentric coordinates of the extracted spine region and the maximum and minimum values of the horizontal axis coordinates of the region including the arbitrary point extracted from the selected tomographic image;
Range determination means for determining whether or not the horizontal axis coordinate of the barycentric coordinate of the spinal region belongs to a range larger than the minimum value of the coordinate in the horizontal axis direction of the region including the arbitrary point and smaller than the maximum value;
Warning means for giving a warning when it is determined by the range determination means that the horizontal axis coordinate of the barycentric coordinate of the spine region does not belong to the range;
The medical image display apparatus according to claim 3, further comprising: The validity determination means determines validity based on a comparison result between a density value of a region including the arbitrary point extracted from the selected tomographic image and a density value of the target organ.
The medical image display apparatus according to claim 3. The display means displays the plurality of tomographic images;
The specifying means is a selecting means for selecting a tomographic image for starting extraction processing of the target organ region from the displayed tomographic images;
Contour line input means for inputting a contour line of the target organ region in the selected tomographic image;
A lock means for changing a density value of a region other than the target organ region out of the selected tomographic image to a density value different from the density value of the target organ region;
The first extracting means extracts the inner area of the contour line after the density value of the other area is changed by the locking means.
The medical image display apparatus according to claim 1. The display means displays the plurality of tomographic images;
The specifying means is a selecting means for selecting a tomographic image for starting extraction processing of the target organ region from the displayed tomographic images;
Correction means for correcting the contour line of the internal region extracted by the first extraction means,
The display means displays the extracted internal region;
The correction means re-extracts the internal region of the contour line corrected on the displayed internal region as the target organ region;
The medical image display apparatus according to claim 1. Display means for displaying the tomographic image constituting a volume image obtained by stacking a plurality of tomographic images obtained by imaging a target organ of a subject;
Selecting means for selecting a plurality of starting tomographic images for starting extraction processing of the target organ region from the displayed plurality of tomographic images;
An input means for inputting a point included in the target organ region on each of the selected starting tomographic images as a determination point;
In the volume image, a line connecting the determination points set on the plurality of starting tomographic images is provisionally set, and an angle formed by the temporarily set line and a plane including the tomographic image is included in the volume image. Determination line setting means for setting a line within a predetermined angle range for passing through the inside of the target organ region as a determination line;
For each tomographic image, candidate region extraction means for performing threshold processing based on the concentration value of the target organ and extracting at least one region including the target organ region as a candidate region;
A target organ region extracting means for extracting, as the target organ region, one that intersects the determination line among the extracted candidate regions,
The display means further displays the extracted target organ region;
A medical image display device characterized by that. A template showing the shape of the target organ region;
The display means displays the template and the starting tomographic image selected by the selection means in an overlapping manner,
The input means inputs the determination point by setting a relative coordinate based on the template.
The medical image display apparatus according to claim 8.

Images