JP4147590B2 - Medical image diagnosis support method, medical image diagnosis support program, and medical image diagnosis support device - Google Patents

Description

The present invention relates to a medical image diagnosis support method , a medical image diagnosis support program, and a medical image diagnosis support device, and more particularly to a method , program, and apparatus for specifying a CT image taken at a navel position.

CT images taken at the navel using an X-ray CT (computed tomography) apparatus are used for measuring the body fat area and for measuring the psoas muscle area. The fat area of the CT image taken at the navel position is used as a standard representing the total fat area of the individual. Patent Document 1 discloses a technique for calculating body fat based on a fat region of a CT image taken at a navel position.
JP 2002-222410 A

  However, until now, there has been no means for automatically extracting a tomographic image at the navel position by a computer or for automatically confirming it.

  For this reason, the operator has to determine whether or not the CT image is taken at the navel position by looking at the CT image, and there is a problem that inspection efficiency is poor.

  The present invention has been made in view of the above problems, and provides a medical image diagnosis support method and apparatus capable of automatically taking out or confirming a tomographic image taken at the navel position with a computer. The purpose is to do.

In order to achieve the object, a medical image diagnosis support method according to the present invention, a medical image diagnosis assisting apparatus, a step of reading a tomographic image of the subject taken by the medical imaging apparatus, the medical image diagnosis assisting apparatus, an extraction step of extracting body tissue area of the subject from the tomographic image, the medical image diagnosis assisting apparatus, a detection step of detecting a recessed area of the extracted body tissue region in the extraction process, the medical image diagnosis assisting apparatus And a sorting step of sorting out the recessed areas close to the navel shape of the subject among the recessed areas detected in the detecting step according to the circularity of the recessed areas. In the detection step, the pixel points on the contour line of the body tissue region extracted in the extraction step are connected with a straight line, and a region where the straight line passes outside the body tissue region is detected as the recess region. May be. Further, in the sorting step, the circularity is calculated based on an aspect ratio of the recessed area, and the recessed area whose circularity is closest to a predetermined value is selected as a recessed area close to the navel shape. Good.

  Furthermore, a determination step of determining whether or not the recessed region selected in the selection step is within a predetermined range in which the navel of the subject is located in the body tissue region may be included.

Further, in the determination step, the body tissue region is divided into a predetermined number of divided regions, based on the relative positional relationship between the body tissue region and the divided regions, it is assumed to include navel of the subject Ruheso divided region is determined, the sorted recessed region in a selection step may be determined based on whether or not included in the navel divided regions.

Further, in the determination step , an angle formed by a reference line passing through the center of gravity of the body tissue region and a straight line passing through the center of gravity of the body tissue region and the pixel points constituting the recessed region is within a predetermined angle range. You may determine based on whether it is contained.

In the determining step , an angle formed by a reference line passing through the center of gravity of the spinal cord region included in the body tissue region and a straight line passing through the center of gravity of the spinal cord region and the pixel points constituting the recessed portion is a predetermined value. You may discriminate | determine based on whether it is contained in an angle range.

Also, if the sorted recessed region in the previous SL sorting process, navel of the subject in the body tissue region is determined in the determination step not within the predetermined range which is assumed to be located, the tomographic image May further include a display step of displaying that the image is not taken at the navel position of the subject.

Further, in the prior SL determination step, when said recessed area is navel of the subject in the tomographic image is determined to not within the predetermined range which is assumed to be located, in the read process, the other tomographic image more read- rare, in the extraction step and the detecting step and the sorting step the determination step, the process we row in each step with respect to read- Mareta the other tomographic image, the read step, in the determination step until it is determined that the recessed area is within a predetermined range navel of the subject is assumed to be located in the tomographic image may be Repetitive Rikae.
Furthermore, in order to achieve the object, a medical image diagnosis support program according to the present invention causes a computer to execute the medical image diagnosis support method according to any one of claims 1 to 7.

In addition, the medical image diagnosis support apparatus according to the present invention includes an input unit that reads a tomographic image of a subject imaged by the medical image capturing apparatus, extracts a body tissue region of the subject from the tomographic image, and extracts the body tissue region from the tomographic image. Based on the recessed area selected by the data processing means, detecting the recessed area obtained by imaging the recessed portion of the subject, and selecting the recessed area close to the navel shape of the subject among the recessed areas, Means for outputting information as to whether or not the recessed area is a navel area of the subject.
The data processing means may determine whether or not the selected recessed area is within a predetermined range in which the subject's navel is located in the body tissue area.
Further, it is assumed that the data processing unit divides the body tissue region into a predetermined number of divided regions and includes a navel of the subject based on a relative positional relationship between the body tissue region and the divided regions. A navel division area to be determined may be determined, and whether or not the concave area is within the predetermined range may be determined based on whether or not the selected concave area is included in the navel division area. .
The data processing means may be configured such that an angle formed by a reference line passing through the center of gravity of the body tissue region and a straight line passing through the center of gravity of the body tissue region and the pixel points constituting the recessed region is within a predetermined angle range. It may be determined whether or not the recessed area is within the predetermined range based on whether or not it is included.
Further, the data processing means has an angle formed by a reference line passing through the center of gravity of the spinal cord region included in the body tissue region and a straight line passing through the center of gravity of the spinal cord region and the pixel points constituting the recessed region. It may be determined whether or not the recessed area is within the predetermined range based on whether or not the angle is included in the angle range.
Further, the means for outputting the information may determine that the tomographic image is obtained when it is determined that the selected recessed region is not within a predetermined range in which the navel of the subject is assumed to be located in the body tissue region. You may display that it was not image | photographed in the navel position of the subject.
The input means further reads another tomographic image when it is determined that the recessed area is not within a predetermined range in which the navel of the subject is located in the tomographic image, and the data processing means And the means for outputting the information also process the other tomographic image that has been read, and the input means has a predetermined region in which the concave region is assumed to be located at the navel of the subject in the tomographic image. The reading of other tomographic images may be repeated until it is determined that they are within the range.

  According to the present invention, the concave portion closest to the navel shape of the subject is selected from the read tomographic image, and it is determined whether or not the concave portion is a navel region. Accordingly, it is possible to provide a medical image diagnosis support method and apparatus capable of automatically taking out or automatically confirming a tomogram at the navel position by a computer.

  Hereinafter, preferred embodiments of a medical image diagnosis support apparatus and method according to the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a hardware configuration diagram showing the configuration of the medical image diagnosis support apparatus according to the first embodiment. FIG. 2 is a schematic diagram showing a configuration of a CT image stored in the image database shown in FIG.

  A medical image diagnosis support system 1 in FIG. 1 includes a medical image capturing apparatus 2 that captures a tomographic image of a subject, an image database 3 that stores a tomographic image captured by the medical image capturing apparatus 2, and a navel based on the tomographic image. A medical image diagnosis support apparatus 10 that performs image processing for detecting a position is provided. The medical image photographing apparatus 2, the image database 3, and the medical image diagnosis support apparatus 10 are connected to a network such as a LAN 4.

  The medical image capturing apparatus 2 is configured by an apparatus capable of capturing a tomographic image of a subject such as an MR apparatus and a PET apparatus in addition to an X-ray CT apparatus. In the present embodiment, an X-ray CT apparatus is used as the medical image photographing apparatus 2 to generate a tomographic image (hereinafter referred to as “CT image”).

  The medical image diagnosis support 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. A system (OS), a device drive of a peripheral device, a magnetic disk 13 storing various application software including a program for specifying a CT image including a navel position described later, and a display memory 14 for temporarily storing display data A monitor 15 such as a CRT monitor or a liquid crystal monitor that displays an image based on data from the display memory 14, a keyboard 16, a mouse 17 as a position input device, and a state of the mouse 17 by detecting the state of the mouse 17. Signals such as the position of the upper mouse pointer and the state of the mouse 17 are output to the CPU 11. A controller 18, a bus 19 for connecting the above components.

  The image database 3 stores CT images captured by the medical image capturing apparatus 2. In chest examination, about 30 CT images are stored from the chest to the abdomen of one subject. These CT images have the same orientation so that the front direction (abdomen direction) of the subject is upward on the screen of the monitor 15 and the rear direction (back direction) of the subject is downward on the screen of the monitor 15. Stored. In FIG. 2A, n CT images are stored from the chest toward the abdomen, and stored in the order from the first to the nth. The nth CT image is a CT image reconstructed at the navel position. In FIG. 2 (b), as in FIG. 2 (a), n CT images are stored from the chest toward the abdomen, but the CT image n reconstructed at the navel position is stored at the top, followed by CT image 1, CT image 2,..., CT image n-2, CT image n-1 are stored in this order. The medical image diagnosis support apparatus 10 reads the stored CT images in order from the top.

  In the present embodiment, the medical image diagnosis support apparatus 10 reads out CT images from the image database 3 via the LAN 4, but a storage device, FDD, CD-RW drive, MO connected to the medical image diagnosis support apparatus 10. A drive, a ZIP drive, and the like may be provided, and image data may be read from these drives. Alternatively, it may be acquired from the medical image photographing apparatus 2 via the LAN 4.

  Next, a procedure for specifying a CT image reconstructed at the navel position will be described with reference to FIGS. FIG. 3 is a flowchart for explaining the medical image diagnosis support method according to the first embodiment, and shows a procedure for specifying a CT image reconstructed at the navel position. FIG. 4 is a schematic diagram illustrating processing for detecting a recess from a CT image reconstructed at the navel position and processing for selecting a recess that is estimated to be a navel region. 5 and 6 are schematic diagrams illustrating a method of determining whether or not a recess that is estimated to be a navel region is included in a predetermined range in the processing target region. Hereinafter, it demonstrates along each step of FIG.

(Step S301)
The medical image diagnosis support apparatus 10 reads one CT image 20 of the subject stored in the image database 3 (S301).

(Step S302)
The medical image diagnosis support apparatus 10 extracts the body tissue region of the subject from the read CT image (S302).

  4 includes an air region 41 having a CT value of approximately −1000, a soft tissue region 42 of the body having a CT value of approximately 0 to 100, and a spinal cord region 43 having a CT value of 2000. The CT image 40 is configured as a binary image in which a region having a CT value of less than 0 is represented by a pixel value “0” and a region having a CT value of 0 or more is represented by a pixel value “1”. The CT image 40 is scanned from the upper left corner to the lower right corner to extract a region having a pixel value of “1”, thereby extracting a body tissue region.

(Step S303)
The contour line of the body tissue extracted in step S302 is extracted. Then, the pixel points on the contour line are connected to each other with a straight line, and a concave region is detected by obtaining a region where the straight line passes outside the other tissue region (S303).

  For example, straight lines 45a, 45b, and 45c connecting the pixel points 44 on the contour line are drawn. Since the straight line 45a passes through the soft tissue region 42 and the spinal cord region 43 (hereinafter referred to as “body tissue regions 42, 43”) constituting the body tissue region, the straight line 45a is determined not to be a recessed region. The straight lines 45 b and 45 c pass outside the body tissue regions 42 and 43. Therefore, a region where the pixel points on the straight lines 45b and 45c are collected is detected as a recessed region. In FIG. 4B, the recessed area 46b and the recessed area 46c are detected. The recessed area 46b is a recessed part made of a navel, and the recessed area 46c is a recessed part on the back.

(Step S304)
The recessed area closest to the navel shape of the subject is selected (S304). As a method for selecting the recessed area close to the navel shape, for example, the degree of circularity of the recessed area is examined, and the recessed area having a shape close to a circle is selected as the recessed area estimated to be the navel area.

  For the degree of circularity, first, the number of Pixels constituting the recessed area is calculated based on the equation (1).

[Formula 1]
Pixels = L * S _eff (L: measured value of long axis, S _eff : execution value)
Next, it calculates by calculating | requiring an aspect ratio (ratio of a major axis and a minor axis) based on Formula 2.

[Formula 2]
(Long axis / short axis) = L / S _eff = L ² / Pixels
The numerical value obtained by Equation 2 is 1 for a circle, about 2 for a navel region, and about 3 for a back region. The degree of circularity of the plurality of recessed areas detected in S303 is calculated from Equation 1 and Equation 2, and the recessed area closest to the value obtained by Equation 2 is estimated to be a navel region. Identify the recessed area. In FIG. 4C, the recessed area 46b is selected.

(Step S305)
Based on the positional relationship between the recessed region 46b selected in step S304 and the body tissue regions 42 and 43 obtained in step S302, it is confirmed whether or not the recessed region 46b selected in step S304 is a navel region. Specifically, as shown in FIG. 5 or FIG. 6, it is determined whether or not the extracted recessed area 46b is located in a predetermined range in the body tissue areas 42 and 43 (S305).

  In FIG. 5, the body tissue regions 42 and 43 are divided into two parts in the vertical direction and three parts in the left and right direction, for a total of six parts. Then, it is determined whether or not the recessed region 46b is included in the upper divided region. If the recessed region 46b is included in the upper divided region, the recessed region 46b is determined as the navel region.

  In FIG. 6A, the recessed area 46b passes through the centroid 61 of the body tissue areas 42 and 43 and passes through the straight line 62 parallel to the body thickness direction of the subject and the pixel point 46b1 constituting the centroid 61 and the recessed area 46b. An angle θ1 formed by the straight line 63 is calculated. If the angle θ1 is within a predetermined range, the recessed area 46b is determined as a navel area. Whether or not the concave region 46b is a navel region by the method shown in FIG. 6A even when the spinal cord region 43 is included in the CT image including the navel position or when the spinal cord region 43 is not included. Can be confirmed.

  In FIG. 6 (b), the recessed region 46b constitutes a straight line 65 passing through the center of gravity 64 of the spinal cord region 43 included in the body tissue regions 42 and 43 and parallel to the body thickness direction of the subject, and the center of gravity 64 and the recessed region 46b. An angle θ2 formed by the straight line 66 passing through the pixel point 46b1 is calculated. If the angle θ2 is within a predetermined range, the recessed area 46b is determined as a navel area. When the spinal cord region 43 is included in the CT image including the navel position, it is possible to confirm whether or not the recessed region 46b is a navel region by the method of FIG. 6B.

  In step S305, when the recessed area 46b selected in step S304 is within a predetermined range, the process proceeds to step S306. If the recessed area 46b selected in step S304 is not within the predetermined range, the process returns to step S301, another CT image is read from the image database 3, and the above steps are performed again.

(Step S306)
In step S305, if the recessed area 46b is within a predetermined range, it can be determined that the CT image is a CT image taken at the navel position. Based on the CT image, desired measurement processing, for example, measurement of a body fat region, measurement processing of the psoas muscle, and the like are performed.

  Next, a medical image diagnosis support method according to the second embodiment will be described with reference to FIGS. FIG. 7 is a flowchart for explaining the medical image diagnosis support method according to the second embodiment. As a result of reading a CT image presumed to be taken at the navel position and performing a navel position specifying process, A method for displaying on the monitor 15 that the CT image is not taken at the navel position will be described. FIG. 8 is a schematic diagram showing a screen display example displayed in step S707 of FIG. Hereinafter, description will be made along the order of steps in FIG.

(Step S701)
First, the medical image diagnosis support apparatus 10 reads a CT image estimated to be taken at the navel position (S701).

(Steps S702 to S704)
The body tissue areas 42 and 43 are detected from the CT image read in step S701 (S702), the recessed area is detected (S703), and only the recessed area estimated as the navel area is selected from the recessed areas (S704). ). From step S702 to step S704, processing similar to that from step S302 to step S304 is performed.

(Step S705)
By the same method as Step S304, it is determined whether or not the recessed area selected in Step S704 is within a predetermined range. If the recessed area is within the predetermined range, the process proceeds to step S706. If the recessed area is not within the predetermined range, the process proceeds to step S707 (S705).

(Step S706)
Measurement processing similar to that in step S306 is performed (S706). Further, a message indicating that the image includes the navel area may be displayed.

(Step S707)
An error message for notifying that the read CT image is not taken at the navel position is displayed (S707). For example, an error message 80 shown in FIG. 8 is displayed on the monitor 15. And it ends.

  In the above embodiment, the body tissue regions 42 and 43 are divided into six in FIG. 5, but the number of divisions is not limited to six.

  In FIG. 6A, a straight line parallel to the body thickness direction passing through the center of gravity 61 is used as a reference line passing through the center of gravity 61 of the body tissue regions 42 and 43. A straight line parallel to the width direction, a straight line having an angle of 45 ° with respect to the body thickness direction and the body width direction may be used as the reference line. The same applies to the reference line passing through the center of gravity 64 of the spinal cord region 43 used in FIG.

1 is a hardware configuration diagram showing the configuration of a medical image diagnosis support apparatus according to a first embodiment. Schematic diagram showing the configuration of a CT image stored in the image database shown in FIG. Flowchart for explaining a medical image diagnosis support method according to the first embodiment Schematic diagram showing a process for detecting a recess from a CT image reconstructed at the navel position and a process for selecting a recess estimated to be a navel region The schematic diagram which shows the method of determining whether the recessed part estimated to be a navel area | region is contained in the predetermined range in a process target area | region. The schematic diagram which shows the method of determining whether the recessed part estimated to be a navel area | region is contained in the predetermined range in a process target area | region. Flowchart for explaining a medical image diagnosis support method according to the second embodiment Schematic diagram showing a screen display example displayed in step S707 of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Medical image diagnosis support system, 2 ... Medical image imaging device, 3 ... Image database, 10 ... Medical image diagnosis support device, 11 ... CPU, 12 ... Main memory, 13 ... Magnetic disk, 14 ... Display memory, 15 ... Monitor 16 ... Keyboard, 17 ... Mouse, 18 ... Controller, 19 ... Common bus

Claims (19)

A step in which a medical image diagnosis support apparatus reads a tomographic image of a subject imaged by a medical image imaging apparatus;
A medical image diagnosis support apparatus for extracting a body tissue region of a subject from the tomographic image;
A detection step in which the medical image diagnosis support apparatus detects a recessed region of the body tissue region extracted in the extraction step;
The medical image diagnosis support apparatus selects a recess area close to the navel shape of the subject among the recess areas detected in the detection process, according to the circularity of the recess area;
A medical image diagnosis support method comprising:
The detecting step connects pixel points on the outline of the body tissue region extracted in the extraction step with a straight line, and detects a region where the straight line passes outside the body tissue region as the recessed region. The medical image diagnosis support method according to claim 1. The selecting step is characterized in that the circularity is calculated based on an aspect ratio of the recessed area, and the recessed area whose circularity is closest to a predetermined value is selected as a recessed area close to the navel shape. Item 2. The medical image diagnosis support method according to Item 1. A determination step of determining whether or not the recessed region selected in the selection step is within a predetermined range in which the navel of the subject is assumed to be located in the body tissue region;
  The medical image diagnosis support method according to any one of claims 1 to 3, wherein the medical image diagnosis support method is used. In the determining step, the body tissue region is divided into a predetermined number of divided regions, and a navel that is assumed to include a navel of the subject based on a relative positional relationship between the body tissue region and the divided regions. A divided area is determined, and is determined based on whether or not the recessed area selected in the sorting step is included in the navel divided area.
The medical image diagnosis support method according to claim 4. In the determining step, an angle formed by a reference line passing through the center of gravity of the body tissue region and a straight line passing through the center of gravity of the body tissue region and the pixel points constituting the recessed region is included within a predetermined angle range. Is determined based on whether or not
The medical image diagnosis support method according to claim 4. A step in which a medical image diagnosis support apparatus reads a tomographic image of a subject imaged by a medical image imaging apparatus;
A medical image diagnosis support apparatus for extracting a body tissue region of a subject from the tomographic image;
A detection step in which the medical image diagnosis support apparatus detects a recessed region from the tomographic image;
The medical image diagnosis support apparatus sorts out the recessed area close to the navel shape of the subject among the recessed areas detected in the detecting process, and
A determination step in which the medical image diagnosis support apparatus determines whether or not the concave region selected in the selection step is within a predetermined range in which the navel of the subject is located in the body tissue region; Including
In the determination step, an angle formed by a reference line passing through the center of gravity of the spinal cord region included in the body tissue region and a straight line passing through the center of gravity of the spinal cord region and the pixel points constituting the recessed region is within a predetermined angular range. Discriminated based on whether it is included in,
A medical image diagnosis support method characterized by the above. When it is determined in the determining step that the recessed region selected in the selecting step is not within a predetermined range in which the navel of the subject is located in the body tissue region, the tomographic image is And further including a display step for displaying that the image is not taken at the navel position.
The medical image diagnosis support method according to any one of claims 4 to 7, In the determination step, when it is determined that the recessed area is not within a predetermined range where the navel of the subject in the tomographic image is assumed to be located,
In the reading step, another tomographic image is further read,
In the extraction step, the detection step, the selection step, and the determination step, processing in each step is performed on the read other tomographic image,
The reading step is repeated until it is determined in the determination step that the concave region is within a predetermined range where the subject's navel is located in the tomographic image.
The medical image diagnosis support method according to any one of claims 4 to 8. A medical image diagnosis support program for causing a computer to execute the medical image diagnosis support method according to any one of claims 1 to 9. Input means for reading a tomographic image of a subject imaged by a medical imaging apparatus;
Extracting the body tissue region of the subject from the tomographic image, detecting a recessed region obtained by imaging the recess of the subject from the extracted body tissue region, and forming a recessed portion close to the navel shape of the subject in the recessed region Data processing means for selecting the area according to the circularity of the recessed area;
Means for outputting information as to whether or not the recessed area is a navel area of the subject based on the recessed area selected by the data processing means;
A medical image diagnosis support apparatus comprising: The data processing means includes means for connecting pixel points on a contour line of the body tissue region with a straight line, and detecting a region where the straight line passes outside the body tissue region as the recessed region. Item 12. The medical image diagnosis support apparatus according to Item 11. The data processing means includes means for calculating the circularity based on an aspect ratio of the recessed area, and selecting a recessed area whose circularity is closest to a predetermined value as a recessed area close to the navel shape. The medical image diagnosis support apparatus according to claim 11, wherein the medical image diagnosis support apparatus is a medical image diagnosis support apparatus. The data processing means determines whether or not the selected recessed area is within a predetermined range where the subject's navel is located in the body tissue area.
The medical image diagnosis support apparatus according to claim 11. It is assumed that the data processing means divides the body tissue region into a predetermined number of divided regions and includes a navel of the subject based on a relative positional relationship between the body tissue region and the divided regions. Determining a navel-divided region, and determining whether the recessed region is within the predetermined range based on whether the selected recessed region is included in the navel-divided region;
The medical image diagnosis support apparatus according to claim 14. The data processing means includes an angle formed by a reference line passing through the center of gravity of the body tissue region and a straight line passing through the center of gravity of the body tissue region and the pixel points constituting the recessed region within a predetermined angle range. Based on whether or not the recessed area is within the predetermined range,
The medical image diagnosis support apparatus according to claim 14. Input means for reading a tomographic image of a subject imaged by a medical imaging apparatus;
The body tissue region of the subject is extracted from the tomographic image, and a concave region obtained by photographing the concave portion of the subject is detected from the tomographic image, and the concave region close to the navel shape of the subject is selected from the concave region. Data processing means;
Means for outputting information as to whether or not the recessed area is a navel area of the subject based on the recessed area selected by the data processing means,
The data processing means is configured such that an angle formed by a reference line passing through the center of gravity of the spinal cord region included in the body tissue region and a straight line passing through the center of gravity of the spinal cord region and the pixel points constituting the recessed region is a predetermined angle. Determining whether or not the recessed area is within the predetermined range based on whether or not it is included in a range;
A medical image diagnosis support apparatus characterized by that. The means for outputting the information may determine that the tomogram is the subject when the selected concave region is not within a predetermined range in which the navel of the subject is assumed to be located in the body tissue region. To display that it was not taken at the navel position,
The medical image diagnosis support apparatus according to claim 11, wherein the medical image diagnosis support apparatus is a medical image diagnosis support apparatus. The input means further reads another tomographic image when it is determined that the recessed area is not within a predetermined range where the navel of the subject in the tomographic image is assumed to be located,
The data processing means and the means for outputting the information also process the other tomographic images that have been read,
The input means repeats reading of other tomographic images until it is determined that the recessed area is within a predetermined range where the navel of the subject in the tomographic image is assumed to be located.
The medical image diagnosis support apparatus according to claim 11, wherein the medical image diagnosis support apparatus is a medical image diagnosis support apparatus.

Images