JP2016073679A - Medical image generation device, medical image generation program and medical image generation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a medical image generation device, a medical image generation program and a medical image generation method each of which enables an optimal window width value/level value in an image at an arbitrary b value to be easily set.SOLUTION: A medical image generation device 1 comprises: first calculation means 21 that obtains A, B and B/A through calculation, where A denotes an image signal average of all pixels in multiple medical images acquired by imaging and B denotes an average of image signals of all pixels in a medical images at an arbitrary factor; second calculation means 22 that obtains a window width value P and a window level value Q related to display of the medical images obtained by imaging and by multiplying each of the window width value PAVE and the window level value QAVE at the arbitrary factor by B/A.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a medical image generation technique for generating a medical image, and particularly based on three-dimensional image data obtained by a medical image imaging apparatus such as CT (Computed Tomography) or MRI (Magnetic Resonance Imaging), The present invention relates to a medical image generation apparatus, a medical image generation program, and a medical image generation method that can generate a two-dimensional visible image suitable for diagnosis.

Magnetic resonance tomographic images (MRI) have high spatial resolution and contrast resolution by capturing the behavior of water molecules, and are therefore used as a means for diagnosing lesions in various regions.
In addition, one method of expressing the magnitude of water molecule behavior is “diffusion”, which indicates the degree of freedom of activity. However, an imaging method called “Diffusion Weighted Image” (DWI) that emphasizes this is known. It has been. This is widely used for various diagnoses in addition to detection of cerebral infarction and tumor.
The DWI can depict a region where water molecules are easily diffused or a region where water molecules are difficult to diffuse by setting a b value indicating the intensity of contrast with an MRI apparatus.

On the other hand, the b value has been proposed to be appropriate for each observational tissue such as an organ. In the diagnosis of prostate cancer, the b value should be taken in a state where the b value is set to a value such as 2000. Is well known. It is also well known to take images for each of a plurality of b values, observe changes in signal values of pixels (picture elements: the same applies hereinafter) between the images, and perform diagnosis based on the changes.
Further, in an orthogonal coordinate system with the b value on the horizontal axis and the pixel signal value on the vertical axis, linear interpolation is performed by forming an approximate straight line connecting signal values corresponding to a plurality of b values, There is also known a technique that an image of an arbitrary b value can be obtained (referred to as “Computed-DWI”).

Radiology: Volume261: Number2-November2011 ・ radiology.rsna.org / Matthew D. Blackledge et al.

Since the image of an arbitrary b value interpolated as described above is an extension of linear calculation, the signal distribution for each pixel is broad, and therefore, in an image of an arbitrary b value, There is a possibility that the maximum value, the minimum value, etc. of the signal are greatly different from the maximum value, the minimum value, etc. of the signal in the b-value image plotted for performing linear interpolation.
However, since the initial values of the window width value and window level value used when displaying a grayscale image on a monitor are unknown, they have been plotted for linear interpolation in many cases. The window width value and window level value in the b-value image were used as they were.

For this reason, after obtaining an image having an arbitrary b value, the operator needs to adjust the window width value and the window level value that are easy to see in the image, which imposes an excessive burden on the operator during interpretation. .
The present invention has been made in view of such circumstances, and a medical image generation device, a medical image generation program, and a medical image that can easily set an optimum value of a window width value / level value in an image of an arbitrary b value The object is to provide an image generation method.

In order to achieve the above object, a medical image generation apparatus, a medical image generation program, and a medical image generation method according to the present invention are configured as follows.
That is, the medical image generation apparatus according to the present invention includes a predetermined factor relating to a medical image signal and a change state of medical image data acquired by imaging a signal value corresponding to this factor for a specific part. A medical image generation device for obtaining medical image data in any of the above factors based on a relationship,
A when the average value of the image signals of all the pixels of the plurality of medical images acquired by the imaging is A and the average value of the image signals of all the pixels of the medical images in any of the factors is A, First calculating means for calculating values of B and B / A by calculation;
The window width value related to the display of a plurality of medical images acquired by imaging by multiplying the window width value PAVE and the window level value QAVE by the B / A value for each arbitrary factor. Second computing means for determining P and window level value Q;
It is characterized by comprising.

  In addition, a method for obtaining medical image data with an arbitrary factor based on changes in the plurality of medical image data takes a b value on one axis and a signal value on the other axis to create an orthogonal coordinate system. The plurality of medical image data is plotted on this orthogonal coordinate system, and a linear approximation process is performed for each of the points and an extrapolation process is performed to obtain a linear graph. A method for obtaining medical image data in a factor is preferable.

Further, the plurality of medical images acquired by imaging are obtained by calculation based on each of two or more diffusion weighted images acquired using different b values using an MRI apparatus. A medical image is preferred.
Further, the plurality of medical images acquired by imaging are obtained by calculation based on each of two or more diffusion weighted images acquired using different b values using an MRI apparatus. A medical image is preferred.

The medical image generation program according to the present invention is based on a relationship between a predetermined factor for a specific part and a change mode of a plurality of pieces of medical image data acquired by imaging while changing the factor. A medical image generation program for causing a computer to execute a process for obtaining medical image data in the above factor,
When the average value of the image signals of all the pixels of the plurality of medical images obtained by imaging is A, and the average value of the image signals of all the pixels of the medical image in any of the factors is B by the interpolation calculation. A first calculation step of calculating a B / A value of
The window width value related to the display of a plurality of medical images acquired by imaging by multiplying the window width value PAVE and the window level value QAVE by the B / A value for each arbitrary factor. A second computing step for determining P and window level value Q;
Is executed in the computer.

Further, the medical image generation method according to the present invention is based on a relationship between a predetermined factor and a change aspect of a plurality of medical image data acquired by imaging while changing this factor for a specific part. A medical image generation method for obtaining medical image data in the factor of
When the average value of the image signals of all the pixels of the plurality of medical images obtained by imaging is A, and the average value of the image signals of all the pixels of the medical image in any of the factors is B by the interpolation calculation. A first calculation step of calculating a B / A value of
The window width value related to the display of a plurality of medical images acquired by imaging by multiplying the window width value PAVE and the window level value QAVE by the B / A value for each arbitrary factor. A second computing step for determining P and window level value Q;
Are performed in this order.

According to the medical image generation device, the medical image generation program, and the medical image generation method according to the present invention,
In a desired medical image, it is possible to quantitatively determine an appropriate window width and window level that can be displayed on a monitor so that diagnosis is good.

  Further, conventionally, when a medical image is displayed on a monitor, the operator sequentially adjusts the window width value and the window level value according to each medical image. However, according to the present invention, all the pixels of the medical image in an arbitrary factor obtained by the interpolation operation with respect to the image signal average value A of all the pixels for a plurality of medical images actually obtained by imaging. Since the calculation of multiplying the ratio of the image signal average value B to each of the window width value P and the window level value Q related to the display of a plurality of medical images acquired by imaging is performed uniformly, automation is easy. Thus, the burden on the operator can be greatly reduced.

  In addition, the DWI image obtained by the MRI apparatus can be easily and accurately displayed on the monitor even in a region where quantitative image display is difficult.

1 is a block diagram illustrating a schematic configuration of a medical image generation apparatus according to an embodiment of the present invention. (A) is obtained based on two diffusion weighted images (DWI) imaged by an MRI apparatus using different b values, and a signal value related to an arbitrary b value can be obtained. A straight line graph is shown, and (B) is obtained based on a diffusion weighted image (DWI) acquired using one b value and an apparent diffusion coefficient image (ADCmap). The linear graph which can obtain | require the signal value which concerns on arbitrary b values is shown. It is a flowchart for demonstrating the flow of a process of the medical image generation method which concerns on one Embodiment of this invention. When the b value is scanned in the increasing direction and the b value is set to 0, the window width value and the window level value are each 2000 so that the medical image captured by the MRI apparatus can be displayed satisfactorily. And the display image (A) when set to 1000, the image data of this display image (A) and the interpolation method shown in FIG. When the display image (B) when the level value is set to 232 and the window width value 2000 and the window level value 1000 set in the display image (A) are used as they are even when the b value is 2000 The display images (C) are respectively shown. When the b value is scanned in the decreasing direction and the b value is 2000, the window width value and the window level value are each 500 so that the medical image captured by the MRI apparatus is displayed well. And the display image (A) when set to 250, the image data of the display image (A) and the interpolation method shown in FIG. When the display image (B) when the level value 1073 is set and the window width value 500 and the window level value 250 set in the display image (A) are used as they are even when the b value is 0 The display images (C) are respectively shown.

Hereinafter, embodiments of the present invention will be described in detail with reference to the above-mentioned drawings.
First, an apparatus configuration of a medical image generation apparatus 1 according to an embodiment of the present invention will be described with reference to FIG.

  The medical image generation apparatus 1 shown in FIG. 1 specifies the biological tissue specified by the operation device 3 including a mouse and a keyboard based on the three-dimensional image data of the biological tissue obtained by the image capturing apparatus (MRI apparatus) 4. This is an apparatus for obtaining medical image data of an arbitrary factor by interpolation based on the relationship between a change in a predetermined factor and a change in a plurality of medical image data acquired by imaging with respect to a part.

As shown in FIG. 1, the apparatus includes a control unit 11, a captured image storage unit 12, an interpolation image storage unit 13, a window width value / level value storage unit 14, a display image output interface 15, an operation input interface 16, and image data. An input interface 17 is provided.
Each unit configured as described above includes hardware such as a CPU and a memory, and software including a program.

  The control unit 11 includes factor (b value) changing means 20, first calculating means 21, second calculating means 22, and window width value / level value setting means 23 (details will be described later).

  The captured image storage unit 12, the interpolated image storage unit 13, and the window width value / level value storage unit 14 are provided so that various data can be transmitted to and received from the control unit 11.

  The captured image storage unit 12 stores three-dimensional image data (MRI image data or the like) of a living tissue obtained by tomography with an image imaging device (MRI apparatus or the like) 4. The interpolated image storage unit 13 stores interpolated image data generated by the control unit 11 using interpolation processing. Further, the window width value / level value storage unit 14 stores the window width value and the window level value obtained by the calculation by the control unit 11.

  The display image output interface 15 is an interface that outputs the two-dimensional display image (two-dimensional display image data) output from the control unit 11 to the image display device (monitor) 2. The operation input interface 16 is an interface that transmits various operation inputs from the operation device 3 to the control unit 11. Further, the image data input interface 17 is an interface for transmitting the three-dimensional image data output from the image capturing apparatus (MRI apparatus or the like) 4 to the control unit 11.

Hereinafter, each means 20-23 in the control part 11 is explained in full detail.
First, the factor changing means 20 is a means for changing the setting of the b value in the imaging apparatus (here, the MRI apparatus).
In addition, the first calculation means 21 calculates, for each of a plurality of medical images obtained by changing the b value, an image signal average value (signal value) A of all pixels and an arbitrary value obtained by interpolation calculation. The image signal average value B of all the pixels of the medical image in the factor and a value B / A obtained by dividing B by A are obtained.

Further, in the XY rectangular coordinate system in which the set b value is taken as the X coordinate and the signal value is taken as the Y coordinate (natural logarithm), the image signal average value (signal) of each medical image obtained for at least two different b values. The values A1, A2, A3) are plotted as R1, R2, R3 as shown in FIG. 2A, and the least square method (other approximation methods can be used instead of the least square method). Are used to find a straight line L that is close to three points of R1 (b value = 100), R2 (b value = 500), and R3 (b value = 1000). Here, by setting the signal values of R1, R2, and R3 to A1, A2, and A3, respectively, the signal value A described above is calculated as an average value of A1, A2, and A3.
The straight line L1 is obtained by the following equation (1).
Sx = S0 · exp (−ADC value × b) (1)
However,
Sx is the signal value of the b value to be obtained.
S0 is a signal value when b value = 0.
The ADC value is the ratio of signal value increase / decrease.
The b value indicates the contrast intensity in the MRI image.

Thereafter, the straight line L1 is extrapolated (dotted line portion in FIG. 2), a point T1 corresponding to b value = 2000 is plotted on the straight line L1, and a signal value B when b value = 2000 is obtained. Find a certain B1.
In this way, the value of B / A is calculated. The calculation process for obtaining the values of A, B, and B / A is performed by the first calculation means 21 as described above.

  Next, the window width value P and the window level value Q immediately before the calculation processing related to the display of a plurality of medical images are multiplied by B / A obtained by the first calculation means 21 to obtain a new window width value W. In addition to setting P · B / A, a new window level value L is set to Q · B / A.

That is, the window width value W and the window level value L are expressed as follows.
W = Display window width value of image immediately before calculation process P × (average image signal value immediately after calculation process / average image signal value immediately before calculation process)
L = Display window level value Q of the image immediately before the calculation process Q × (average image signal value immediately after the calculation process / average image signal value immediately before the calculation process)

Thus, the calculation process for obtaining P · B / A and Q · B / A is performed by the second computing means 22.
The image immediately before the calculation process indicates the immediately preceding image for updating the window width and the window level by the calculation process, and the image immediately after the calculation process is the window width and the window level updated by the calculation process. The image at the time is shown.

  When the window width value W and the window level value L are obtained by the second calculating means 22, these values are obtained by the window width value / level value setting means 23 of the control unit 11. In response to the window width value W and the window level value L, the window width value W and the window level value L stored in the window width value / level value storage unit 14 are updated.

Then, using the window width value W and the window level value L, gradation display is set for each pixel of the interpolation image stored in the interpolation image storage unit 13.
Thereafter, the interpolated image that is displayed in shades for each pixel is sent to the image display device (monitor) 2 via the display image output I / F 15 and displayed.

Hereinafter, the effects of the above method will be described.
In the following description, similarly to the above, the image signal average value (signal value) of all pixels of the acquired image is A ′, and the image signal average value obtained by the calculation process is B ′. The display window width value of the image immediately before processing is P ′, the display window level value of the image immediately before calculation processing is Q ′, the window width value after calculation processing is W ′, and the window level value after calculation processing is L ′. To do.

First, when the b value is increased on the semi-logarithmic orthogonal coordinate system graph, the signal value of each pixel decreases. Similarly, the average value in the signal distribution also decreases. The signal drop is exponential and approaches asymptotically. Therefore, the signal value width becomes narrower.
In the present embodiment described above, it is proposed to use the following expressions (2) and (3) after all.
W '= (B' / A '), P' (2)
L '= (B' / A '), Q' (3)

In this case, since the value of (B ′ / A ′) decreases as the b value increases, both W ′ and L ′ can be displayed with a decrease as the b value increases.
Furthermore, by adopting the “average value of signal distribution”, it is possible to obtain a value corresponding to the bias of the histogram of the signal distribution, and even if there is a bias in the data, an appropriate window level value and window width The value can be determined.

  In the above-described embodiment, a plurality of medical images calculated based on two or more diffusion weighted images (DWI) obtained by using an MRI apparatus and using different b values as the plurality of medical images, However, using an MRI apparatus, as shown in FIG. 2B, a diffusion weighted image (DWI) acquired using one b value and an apparent diffusion coefficient image (ADCmap (linear It is also possible to obtain a plurality of medical images calculated based on the inclination α)).

  That is, the points Z at which the image signal average value (signal value) A of the medical image obtained for one b value is 1.0 are plotted as shown in FIG. On the other hand, the straight line L2 having the inclination α (e / d in the figure) obtained from the apparent diffusion coefficient image (ADCmap) is set so as to pass through the point Z. This straight line L2 can also be obtained using the above-described equation (1).

Thereafter, the straight line L2 is extrapolated (dotted line portion in FIG. 2B), a point T2 corresponding to b value = 2000 is plotted on the straight line L2, and the signal value when b value = 2000 is plotted. Find B2 as B.
In this way, B / A is calculated as B2 / 1.0, but the calculation process for obtaining the value of A, the value of B, and B / A is performed by the first computing means (21).
Since the subsequent arithmetic processing is the same as the arithmetic processing of the above embodiment, the description thereof is omitted.

  In the present embodiment, the window width value W and the window level value L are obtained in consideration of the ratio between the image signal average value A before the interpolation calculation and the image signal average value B after the interpolation calculation. The inventors have found that when each point is plotted on a semi-log orthogonal coordinate system of b value-signal value (natural logarithm), the pixel signal value increases or decreases almost linearly with respect to the transition of the b value. It was made based on.

That is, based on the above knowledge, the signal average value of all the pixels of the image is obtained for the pre-interpolation calculation A and the post-interpolation calculation B, and the ratio (B / A) is obtained for each window width of a plurality of medical images. By multiplying the value Pα (α = 1, 2, 3...) And each window level value Qβ (β = 1, 2, 3,...), The window width value P and the window level value Q after interpolation calculation are obtained, respectively. ing.
Thereby, from the initial stage, it is possible to easily set the optimum values of the window width and the window level value as the signal values in an image having an arbitrary b value.

  In the above-described embodiment, the medical image generation technique according to the present invention is applied to “signal transition with respect to the b value in the diffusion weighted image”, but is not limited thereto. The present invention can also be applied to the case of “signal transition with respect to time (time density curve)” in which the signal value of each pixel changes with time.

  In the embodiment of the present invention, the new window width and window level are calculated using a graph, but may be calculated using mathematical formulas and tables instead of the graph. Even when a graph is employed, various graphs, such as a curve such as a quadratic curve, can be used. Further, in the Cartesian coordinate system, one or both of the horizontal axis and the vertical axis may be used as a logarithmic display axis, and it is of course possible to set neither axis as a logarithmic axis.

Here, based on the flowchart shown in FIG. 3, a medical image generation method according to an embodiment of the present invention will be described.
First, a plurality of MRI captured image data with different b values are acquired for each image, and the acquired captured image data is stored (S1).
Next, data corresponding to each captured image is plotted on a semi-log orthogonal coordinate system of b value-signal value (natural logarithm) to obtain a straight line, and linear interpolation processing is performed using the straight line (S2).
Next, a signal value corresponding to an arbitrary b value (signal value B1 (point T) corresponding to b value = 2000 in the example of FIG. 2A) is calculated using the linear interpolation straight line obtained in S2. (S3).

Next, the value of B / A is calculated (first calculation step (S4)).
Next, a window width value W (= P · B / A) and a window level value L (= Q · B / A) are calculated (second calculation step (S5)).

Thereafter, based on the obtained P · B / A and Q · B / A, the window width value and the window level value related to the image display are set, and the medical image data thus generated is displayed as an image. The data is transmitted to the device (monitor) 2 (S6).
Next, FIG. 4 shows images obtained when the b value is scanned in the increasing direction, where the b value is 0 and the window width value and window level value in each image are set to 2000 and 1000, respectively. The acquired image in is shown.

Among them, FIG. 4A shows that the window width value and the window level value are set to 2000 and 1000, respectively, so that the medical image captured by the MRI apparatus is displayed well when the b value is 0. The display image when it is done is shown.
FIG. 4B shows window width values and window level values obtained when the image data of FIG. 4A and the interpolation method shown in FIG. And the display image when set to 232 is shown.
FIG. 4C shows a display image when the window width and window level values 2000 and 1000 set in FIG. 4A are used as they are even when the b value is 2000. Is shown.

  As is clear from FIGS. 4B and 4C, by using the interpolation method according to this embodiment, a simple and good contrast can be obtained, and the window width value and window level value can be obtained. It is possible to automatically set the optimum value of.

  Next, FIG. 5 shows an image obtained when the b value is scanned in the decreasing direction. When the b value is 2000, the window width value and the window level value in each image are set to 2000 and 1000, respectively. The acquired image is shown. Acquired images when the window width value and window level value in each image are set to 500 and 250, respectively, so that the medical image captured by the MRI apparatus is displayed satisfactorily when the b value is 2000 Is shown.

Among them, FIG. 5A shows that when the b value is 2000, the window width value and the window level value in each image are 500 and so that the medical image captured by the MRI apparatus can be satisfactorily displayed. The display image when set to 250 is shown.
5B uses the image data of FIG. 5A and the interpolation method shown in FIG. 3, and the window width value and window level value obtained when the b value is 0 are 2146 and The display image when set to 1073 is shown.
FIG. 5C shows a display image when the window width value and window level value 500 and 250 set in FIG. 5A are used as they are even when the b value is 0. Is shown.

  As can be seen from FIGS. 5B and 5C, by using the interpolation method according to the present embodiment, a simple and good contrast can be obtained, and the window width value and window level value can be obtained. It is possible to automatically set the optimum value of.

DESCRIPTION OF SYMBOLS 1 Medical image generation apparatus 2 Image display apparatus (monitor)
3 Operating device 4 Imaging device (MRI)
11 Control Unit 12 Captured Image Storage Unit 13 Interpolated Image Storage Unit 14 Window Width Value / Level Value Storage Unit 15 Display Image Output I / F
16 Operation input I / F
17 Image data input I / F
20 factor (b value) changing means 21 first calculating means 22 second calculating means 23 window width value / level value setting means

Claims (6)

Based on the relationship between a predetermined factor related to the medical image signal and a change value of the medical image data obtained by imaging a signal value corresponding to this factor for the specific part, the medical image data in an arbitrary factor A medical image generation device for
A when the average value of the image signals of all the pixels of the plurality of medical images acquired by the imaging is A and the average value of the image signals of all the pixels of the medical images in any of the factors is A, First calculating means for calculating values of B and B / A by calculation;
The window width value related to the display of a plurality of medical images acquired by imaging by multiplying the window width value PAVE and the window level value QAVE by the B / A value for each arbitrary factor. Second computing means for determining P and window level value Q;
A medical image generation apparatus comprising:   Based on the change of the plurality of medical image data, a method for obtaining medical image data in an arbitrary factor takes a b value on one axis and takes a signal value on the other axis to create an orthogonal coordinate system, A plurality of medical image data is plotted on this orthogonal coordinate system, and a straight line approximation process is performed for each point and an extrapolation interpolation process is performed to obtain a straight line graph. The medical image generation apparatus according to claim 1, wherein the medical image generation apparatus is a method for obtaining medical image data.   The plurality of medical images acquired by imaging are medical images obtained by calculation based on each of two or more diffusion weighted images acquired using different b values using an MRI apparatus. The medical image generation apparatus according to claim 1, wherein the medical image generation apparatus is a medical image generation apparatus.   The plurality of medical images acquired by imaging are one diffusion weighted image (DWI) acquired using one b value using an MRI apparatus, and an apparent diffusion coefficient image (ADCmap). The medical image generation apparatus according to claim 1, wherein the medical image generation apparatus includes a plurality of medical images obtained by calculation based on the medical image. For a specific part, based on the relationship between a predetermined factor and a change mode of a plurality of medical image data acquired by imaging while changing this factor, processing for obtaining medical image data in any of the above factors, A medical image generation program to be executed on a computer,
When the average value of the image signals of all the pixels of the plurality of medical images obtained by imaging is A, and the average value of the image signals of all the pixels of the medical image in any of the factors is B by the interpolation calculation. A first calculation step of calculating a B / A value of
The window width value related to the display of a plurality of medical images acquired by imaging by multiplying the window width value PAVE and the window level value QAVE by the B / A value for each arbitrary factor. A second computing step for determining P and window level value Q;
Is executed by the computer. Medical image generation for obtaining medical image data for an arbitrary factor based on a relationship between a predetermined factor and a change mode of a plurality of medical image data acquired by imaging while changing the factor for a specific part A method,
When the average value of the image signals of all the pixels of the plurality of medical images obtained by imaging is A, and the average value of the image signals of all the pixels of the medical image in any of the factors is B by the interpolation calculation. A first calculation step of calculating a B / A value of
The window width value related to the display of a plurality of medical images acquired by imaging by multiplying the window width value PAVE and the window level value QAVE by the B / A value for each arbitrary factor. A second computing step for determining P and window level value Q;
Are performed in this order, and a medical image generation method.