JP6143569B2 - Image processing apparatus and region of interest setting method - Google Patents

Description

  The present invention relates to an image processing apparatus and a region-of-interest setting method, and more particularly to an image processing apparatus including means for setting a region of interest.

  The region of interest (hereinafter abbreviated as ROI) in the image processing apparatus is set on the image displayed on the image display unit in the image processing apparatus, and the average, variance, The signal-noise ratio (hereinafter abbreviated as SNR) is analyzed.

  In some cases, an ROI (hereinafter abbreviated as a reference ROI) is set and used in order to handle an analysis value relative to a ROI to be analyzed (hereinafter abbreviated as an analysis ROI).

  ROI is set by setting an area of a closed curve such as an ellipse or a rectangle after an operator visually confirms a desired location on an image displayed on a user interface (hereinafter abbreviated as UI). And a method in which an image processing apparatus automatically sets using one type of image or a shooting condition (shooting position, shooting angle, image quality condition), etc. relating to the image, as shown in Patent Document 1. Yes.

JP 2000-126150 A

  However, when automatic setting is performed by the method shown in Patent Document 1, information is held on one pixel over two or more images (for example, a magnetic resonance imaging apparatus (hereinafter abbreviated as MRI)). In this case, two or more images such as an amplitude image, a phase image, a multi-echo image, etc.) An ROI is set without obtaining necessary information only with information of one image, and an analysis result with low accuracy There is a possibility of producing.

  Accordingly, the present invention has been made in view of the above problems, and an object of the present invention is to set a reference ROI suitable for analysis when information is held on one pixel over two or more images. And

  In order to solve the above problems, main features of the image processing apparatus according to the present invention are as follows.

  An imaging unit that images a desired region to be inspected arranged in a static magnetic field by nuclear magnetic resonance, a display unit that displays an imaging result by the imaging unit, a storage unit that stores data including the imaging result, and an imaging result An arithmetic processing unit that processes data, and an ROI for analysis is set on the first image captured by the imaging unit displayed on the display unit by the operator, and is stored in the storage unit by the arithmetic processing unit The second image is set based on the image information related to the first image, and the reference ROI that contributes to the image analysis of the analysis ROI is determined and determined on the set second image based on the analysis ROI. The reference ROI is displayed on the first image.

  That is, the present invention automatically extracts the first image displayed on the UI and the reference ROI for the analysis ROI set on the first image using the second image paired with the first image, A reference ROI is automatically set on one image.

  According to the present invention, when information is held on one pixel over two or more images, it is possible to set a reference ROI suitable for analysis, and to set a region of interest analysis with high accuracy. Can be performed.

1 is a block diagram illustrating a schematic configuration of an image processing apparatus according to an embodiment of the present invention. It is a figure which shows the analysis screen structure which performs region of interest setting. It is a flowchart which shows Example 1 of this invention. It is a flowchart which shows the reference ROI setting process of FIG. It is a figure of the 1st image used in Example 1 of this invention. It is the figure which set analysis ROI to the 1st image used in Example 1 of the present invention. It is a figure of the result of having set reference ROI to the 1st picture in Example 1 of the present invention. It is a figure of the 2nd image used in Example 1 of this invention. It is the figure which set analysis ROI to the 2nd image used in Example 1 of the present invention. It is a figure at the time of setting a reference ROI candidate to the 2nd picture in Example 1 of the present invention. It is a flowchart which shows Example 2 of this invention. It is a flowchart which shows the reference ROI setting process of FIG.

In the region of interest analysis when using an image in which information is held over a plurality of images, the present invention automatically uses the second image corresponding to the first image to be analyzed to automatically generate the reference ROI used for the analysis. Set up.
Hereinafter, embodiments of the present invention will be specifically described.

The configuration of the image processing apparatus according to this embodiment will be described with reference to FIGS. 1 and 2.
First, as shown in FIG. 1, the image processing apparatus 1 includes a CPU 2, a main memory 3, a data storage device 4, and a network adapter 10, and further includes a display memory 5, a monitor 6, a controller 7, a mouse 8, and a keyboard 9. . Each unit is connected via a bus (arrow in the figure).
The CPU 2, the main memory 3, the data storage device 4, and the like are connected to the LAN 11 via the network adapter 10, and the image DB 12, the medical image photographing device 13, and the like may be connected to the LAN 11.

  The CPU 2 calls a program stored in the main memory 3 or the data storage device 4 to the work memory area of the main memory 3 and executes the program, drives and controls each unit connected via the bus, and executes various processes.

  The display memory 3 is a buffer that temporarily accumulates display data input from the CPU 2. The accumulated display data is output to a predetermined monitor 6.

  The mouse 8 and the keyboard 9 output various instructions and information input by the operator to the CPU 2.

  The LAN 11 may be another communication network, for example, a communication network such as a WAN (Wide Area Network), an intranet, and the Internet.

  FIG. 2 is a diagram showing a configuration of an analysis screen for setting a region of interest. This analysis screen is displayed on the monitor 6.

  The analysis screen includes an image display unit, a menu bar display unit, an analysis method selection unit, a parameter input unit, an analysis method display unit, an image list display unit, and a start button display unit.

Example 1
Next, Example 1 will be described with reference to the flowcharts of FIGS.
[Step S101]
The operator selects a desired first image and loads it on the analysis screen. In this embodiment, an amplitude image as shown in FIG. 5 is used as the first image.

[Step S102]
The operator selects a desired analysis method for the first image from the analysis method selection unit shown in FIG.
In this embodiment, a case where “susceptibility value analysis” is selected as a desired analysis method will be described as an example.
Magnetic susceptibility value analysis is used in an MRI image, and analyzes the magnetic susceptibility representing the degree of magnetization of a substance for each pixel, displays the ratio with surrounding pixels as a color map, and outputs an absolute value. It is. In the magnetic susceptibility value analysis, a magnetic susceptibility image is calculated using an amplitude image and a phase image obtained by measurement, and the magnetic susceptibility value in each pixel of the magnetic susceptibility image is used as an analysis result.

[Step S103]
The operator selects the shape of the ellipse analysis ROI and sets the analysis ROI on the first image.
As an example of the setting, FIG. 6 shows a state in which the analysis ROI is set on the first image. The set analysis ROI is indicated by a circle in the figure. The shape of the analysis ROI is not limited to an ellipse (including a circle), but may be a closed curve such as a rectangle.

[Step S104]
The operator clicks the start button display part (see FIG. 2) on the analysis screen to start analysis.

[Step S105]
The image processing apparatus 1 executes a second image reading process. A phase image as shown in FIG. 8 is used as the second image used in this embodiment.
The second image is read by the image processing apparatus 1 searching the image DB 12 based on the information of the first image. The information of the first image used for the search may be an imaging time, an image identification number, or the like. The second image is not displayed on the monitor 6 but is processed on the main memory 3.

[Step S106]
The image processing apparatus 1 executes a reference ROI setting process. The reference ROI setting process flow will be described later with reference to FIG.

[Step S107]
The reference ROI set on the first image in step S119 of FIG. 4 to be described later is displayed on the image on the monitor 6. At this time, it is displayed on the monitor 6 as shown in FIG. The reference ROI is shown as a small circle in the figure. A large circle indicates the analysis ROI described above.

[Step S108]
Using the analysis ROI set in step S103 and the reference ROI set in step S107, the analysis is performed by the analysis method selected in step S102.

As described in step S102 above, in this embodiment, magnetic susceptibility value analysis is selected as the analysis method. Therefore, the magnetic susceptibility value analysis will be described in detail below.
A magnetic susceptibility value at each pixel is calculated from the amplitude image that is the first image and the phase image that is the second image, and a magnetic susceptibility image is created. At this time, the method for obtaining the magnetic susceptibility is not particularly limited.

  In the susceptibility image, an average susceptibility value in the reference ROI set in step 106 is calculated. This is the reference magnetic susceptibility value. In this embodiment, the average magnetic susceptibility value is used as the reference magnetic susceptibility value, but the maximum magnetic susceptibility value, the minimum magnetic susceptibility value, and the like can also be used as the reference magnetic susceptibility values, but this is not restrictive.

  Next, in each of the pixels in the analysis ROI set in step 103 in the susceptibility image, a ratio with the reference susceptibility value is calculated, and this is used as an analysis value.

  The analysis value obtained as described above is displayed as a color map on the amplitude image which is the first image. For example, a certain threshold value is provided, and a pixel having an analysis value equal to or greater than the threshold value is displayed in red, and a pixel having an analysis value less than the threshold value is displayed in green as a two-stage color map.

  Further, the threshold value at this time may be calculated from an analysis value in the analysis ROI, may be determined in advance, or may be determined by an operator.

≪Reference ROI setting process flow≫
Hereinafter, the reference ROI setting process executed in step 106 will be described in detail with reference to FIG.
[Step S111]
An analysis ROI that is the same position as the analysis ROI set on the first image and has the same size is set on the second image. A state in which the analysis ROI is set on the second image is shown in FIG. 9 (circle in the figure).

[Step S112]
In the range of the analysis ROI set on the second image, a differential filter is applied to calculate an inclination value. As the differential filter used here, a desired filter such as a Sobel filter or a Laplacian filter can be used.

[Step S113]
The maximum inclination value is calculated from the inclination value calculated in step S112. The calculated maximum inclination value is stored in the main memory 3. In addition to the maximum inclination value, the value calculated here may be a center inclination value or an average inclination value.

[Step S114]
The size of the reference ROI to be automatically set is determined from the size of the analysis ROI. In this embodiment, 30% of the major and minor axes of the ellipse are set as the major and minor axis sizes of the elliptical reference ROI, respectively, but the ratio is previously held by the image processing apparatus 1. Or a desired value set by the operator.

[Step S115]
Next, a position that is a candidate for the reference ROI is searched. In this embodiment, the entire image is searched, but the background other than the background on the image may be extracted, and the search may be performed within that range.

[Step S116]
A differential filter is applied to all the searched reference ROI candidates, and a slope value is calculated. The differential filter used here must have the same shape as the differential filter used in step S112.

[Step S117]
Using the inclination value of the reference ROI candidate calculated in step S116, the maximum inclination value of each reference ROI candidate is calculated. The value calculated here may be a center inclination value or an average inclination value in addition to the maximum inclination value, but it should be the same value as that obtained in step S113.

[Step S118]
One of the maximum slope values of each reference ROI candidate calculated in step S117 is determined as a reference ROI that is the minimum and not more than the maximum slope value of the analysis ROI calculated in step S113. At this time, when a plurality of reference ROI candidates correspond to the above condition, in this embodiment, the one having the shortest distance from the center of the analysis ROI to the center of the reference ROI candidate is selected, but this is not restrictive. FIG. 10 shows a state when there are a plurality of reference ROI candidates. In the figure, a plurality of reference ROI candidates are illustrated in a small circle. The large circle is an analysis ROI.

  If there is no one that satisfies the above condition, one reference ROI candidate having the smallest maximum slope value is determined as a reference ROI.

[Step S119]
A reference ROI having the position and size determined in step S118 is set on the first image.
In the present embodiment, the case where “susceptibility value analysis” is selected as the analysis method is shown, but the analysis method is not limited to this. As other analysis methods, SNR (signal-noise ratio) or amplitude value analysis may be used.

  According to the present embodiment, when information is held on one pixel over two or more types of images, it is possible to set and perform a region of interest analysis with high accuracy.

  Further, the obtained analysis value is displayed as a color map on the amplitude image, which is the first image, so that the user can visually easily analyze the region of interest.

(Example 2)
Next, Embodiment 2 will be described with reference to the flowcharts of FIGS.
[Step S201]
The operator selects a desired first image, loads it on the analysis screen, and displays it on the monitor 6. In this embodiment, as in the first embodiment, an amplitude image as shown in FIG. 5 is used.

[Step S202]
The operator selects a desired second image and loads it on the analysis screen. In the present embodiment, a phase image as shown in FIG. 8 is used as in the first embodiment.

[Step S203]
The operator selects a desired analysis method for the first image from the analysis method selection unit shown in FIG.

  In this embodiment, a case where “signal intensity ratio analysis” is selected as a desired analysis method will be described as an example.

  Signal intensity ratio analysis is an analysis method that displays the pixel value of each pixel as a signal value, the signal ratio with surrounding pixels as a color map, and outputs the distribution of signal values. In the signal intensity ratio analysis, the pixel value in each pixel of the amplitude image obtained by measurement is used as the analysis result.

[Step S204]
The operator selects the shape of the ellipse analysis ROI and sets the analysis ROI on the first image.
As an example of the setting, FIG. 6 shows a state in which the analysis ROI is set on the first image, as in the first embodiment. The shape of the analysis ROI is not limited to an ellipse (including a circle), but may be a closed curve such as a rectangle.

[Step S205]
The operator clicks the start button display part (see FIG. 2) on the analysis screen to start analysis.

[Step S206]
The image processing apparatus 1 executes a reference ROI setting process. The reference ROI setting process of this embodiment is shown in FIG. The reference ROI setting process flow will be described later with reference to FIG. 12 [Step S207].
The reference ROI set on the first image in step S217 is displayed on the image on the monitor 6.

[Step S208]
Using the analysis ROI set in step S204 and the reference ROI set in step S207, the analysis is performed by the analysis method selected in step S102.
As described in step S203 above, in this embodiment, the signal intensity ratio analysis is selected as the analysis method. Therefore, the signal intensity ratio analysis will be described in detail below.

  In the amplitude image that is the first image, an average pixel value in the reference ROI is calculated. This is the reference pixel value. In this embodiment, the average pixel value is used as the reference pixel value, but the maximum pixel value or the minimum pixel value may be used as the reference pixel value.

  Next, within the range of the analysis ROI in the amplitude image, the ratio between the pixel value of each pixel and the reference pixel value is calculated and used as the analysis value.

  The analysis value obtained as described above is displayed as a color map on the amplitude image which is the first image. For example, a certain threshold A and B are provided, a pixel having an analysis value greater than or equal to the threshold A is red, a pixel having an analysis value less than or equal to the threshold A is green, and a pixel having an analysis value less than the threshold B is blue. Displayed as a three-level color map.

  Further, the threshold values A and B at this time may be calculated from the analysis value in the analysis ROI, may be determined in advance, or may be determined by the operator.

≪Reference ROI setting process flow≫
[Step S211]
An analysis ROI having the same position and the same size as the analysis ROI set in step S204 is set on the second image.

[Step S212]
Within the analysis ROI set in step S211, the variance value, maximum value, minimum value, and average value of the pixel values are calculated and stored in the main memory 3.

上記縮小率と、解析ＲＯＩの長軸または短軸の大きさを掛け合わせた値を、参照ＲＯＩの長軸または短軸の大きさとする。 [Step S213]
The reference ROI size is determined using the maximum value and the average value calculated in step S212. In this embodiment, since the reference ROI is elliptical, the reduction ratios of the long axis and the short axis are calculated from the maximum value and the average value calculated in step S212. The reduction ratio Ratio is calculated by the following (Expression 1), with the maximum value being Max and the average value being Ave.

A value obtained by multiplying the reduction ratio by the size of the long axis or short axis of the analysis ROI is set as the size of the long axis or short axis of the reference ROI.

[Step S214]
Search for a candidate position for the reference ROI. In this embodiment, the entire image is searched, but the background other than the background on the image may be extracted, and the search may be performed within that range.

[Step S215]
The variance value, maximum value, and minimum value of the pixel value are calculated for all the searched reference ROI candidates.

[Step S216]
A reference ROI is determined such that the maximum value and the minimum value calculated in step S215 are minimum and the variance value calculated in step S215 is smaller than the variance value of the analysis ROI calculated in step S212. At this time, when there are a plurality of conditions that satisfy the above conditions, in this embodiment, the one with the smallest variance value is selected, but this is not restrictive.

[Step S217]
The reference ROI having the position and size determined in step S216 is set on the first image.
The present embodiment can be expected to be particularly effective when applied to multi-echo.

(Example 3)
Next, Example 3 of the present embodiment will be described. The present embodiment is an embodiment when there are a plurality of reference ROI candidates that meet the conditions in step S118 of the first embodiment and step S216 of the second embodiment.

[Step S216 ′]
A reference ROI is determined such that the maximum value and the minimum value calculated in step S215 are minimum and the variance value calculated in step S215 is smaller than the variance value of the analysis ROI calculated in step S212. At this time, when there are a plurality of conditions that satisfy the above condition, in this embodiment, a reference ROI candidate corresponding to the above condition is set and displayed on the first image displayed on the monitor 6.

[Step S217 ′]
The operator selects a desired reference ROI candidate from the displayed reference ROI candidates and determines it as a reference ROI.

[Step S207 ′]
Using the analysis ROI set in step S204 and the reference ROI set in step S217 ′, the analysis is performed by the analysis method selected in step S102.

  According to the present embodiment, for example, it can be expected that the user can select one of the reference ROIs outside the image from the plurality of reference ROIs.

DESCRIPTION OF SYMBOLS 1 ... Image processing apparatus, 2 ... CPU, 3 ... Main memory, 4 ... Data storage device, 5 ... Display memory, 6 ... Monitor, 7 ... Controller, 8 ... Mouse, 9 ... Keyboard, 10 ... Network adapter, 11 ... LAN , 12 ... Image DB, 13 ... Medical imaging device.

Claims (10)

An imaging unit for imaging a desired region to be inspected arranged in a static magnetic field by nuclear magnetic resonance; a display unit for displaying an imaging result by the imaging unit; a storage unit for storing data including the imaging result; An arithmetic processing unit that processes data of the imaging result,
The operator sets an analysis ROI on the amplitude image captured by the imaging unit displayed on the display unit,
A phase image is set by the arithmetic processing unit based on the image information related to the amplitude image stored in the storage unit, and the analysis ROI of the analysis ROI is set based on the analysis ROI on the set phase image. A reference ROI that contributes to image analysis is determined,
Display reference ROI said determined on the amplitude image,
An image processing apparatus , wherein a plurality of reference ROI candidates are set on the phase image based on the analysis ROI, and the reference ROI is determined using a slope of each ROI . The image processing apparatus according to claim 1, wherein the image analysis is performed using analysis values calculated by the analysis ROI and the reference ROI. An imaging unit for imaging a desired region to be inspected arranged in a static magnetic field by nuclear magnetic resonance; a display unit for displaying an imaging result by the imaging unit; a storage unit for storing data including the imaging result; An arithmetic processing unit that processes data of the imaging result,
The operator sets an analysis ROI on the amplitude image captured by the imaging unit displayed on the display unit,
A phase image is set by the arithmetic processing unit based on the image information related to the amplitude image stored in the storage unit, and the analysis ROI of the analysis ROI is set based on the analysis ROI on the set phase image. A reference ROI that contributes to image analysis is determined,
Displaying the determined ROI for reference on the amplitude image;
The reference ROI is an image processing apparatus characterized by being determined using variance of the analysis in the ROI, the maximum and minimum values. The image processing apparatus according to claim 3 , wherein the image analysis is performed using analysis values calculated by the analysis ROI and the reference ROI . An imaging unit for imaging a desired region to be inspected arranged in a static magnetic field by nuclear magnetic resonance; a display unit for displaying an imaging result by the imaging unit; a storage unit for storing data including the imaging result; A region-of-interest setting method in an image processing apparatus comprising an arithmetic processing unit that processes imaging result data,
A step of setting an ROI for analysis on an amplitude image captured by the imaging unit displayed on the display unit by an operator;
A step of setting a phase image based on image information related to the amplitude image stored in the storage unit by the arithmetic processing unit;
Determining a reference ROI that contributes to image analysis of the analysis ROI on the phase image based on the analysis ROI;
Displaying the determined reference ROI on the amplitude image;
A region-of-interest setting method , wherein a plurality of reference ROI candidates are set on the phase image based on the analysis ROI, and the reference ROI is determined using a slope of each ROI . The region- of- interest setting method according to claim 5 , wherein the image analysis is performed using analysis values calculated by the analysis ROI and the reference ROI . An imaging unit for imaging a desired region to be inspected arranged in a static magnetic field by nuclear magnetic resonance; a display unit for displaying an imaging result by the imaging unit; a storage unit for storing data including the imaging result; A region-of-interest setting method in an image processing apparatus comprising an arithmetic processing unit that processes imaging result data,
A step of setting an ROI for analysis on an amplitude image captured by the imaging unit displayed on the display unit by an operator;
A step of setting a phase image based on image information related to the amplitude image stored in the storage unit by the arithmetic processing unit;
Determining a reference ROI that contributes to image analysis of the analysis ROI on the phase image based on the analysis ROI;
Displaying the determined reference ROI on the amplitude image;
The region-of- interest setting method , wherein the reference ROI is determined using a variance value, a maximum value, and a minimum value in the analysis ROI . The region-of-interest setting method according to claim 7 , wherein the image analysis is performed using analysis values calculated by the analysis ROI and the reference ROI . An imaging unit for imaging a desired region to be inspected arranged in a static magnetic field by nuclear magnetic resonance; a display unit for displaying an imaging result by the imaging unit; a storage unit for storing data including the imaging result; A region-of-interest setting method in an image processing apparatus comprising an arithmetic processing unit that processes imaging result data,
A step of setting an ROI for analysis on an amplitude image captured by the imaging unit displayed on the display unit by an operator;
A step of setting a phase image based on image information related to the amplitude image stored in the storage unit by the arithmetic processing unit;
Determining a reference ROI that contributes to image analysis of the analysis ROI on the phase image based on the analysis ROI;
Displaying the determined reference ROI on the amplitude image;
Calculating a magnetic susceptibility value at each pixel of the amplitude image and the phase image;
Creating a magnetic susceptibility image based on the magnetic susceptibility value;
Calculating an average susceptibility value in the reference ROI in the susceptibility image;
Calculating each ratio of the average susceptibility value for each pixel of the susceptibility image in the analysis ROI set by the operator, and setting the calculated ratio as an analysis value;
ROI setting method characterized by having a step of displaying the analysis values, on the amplitude image. The region-of-interest setting method according to claim 9 , wherein a predetermined threshold value is provided, the analysis values are grouped based on the threshold value, and the analysis values are color-coded according to an identification color for each group .

Images