JP7364227B2 - Breast phantom manufacturing method, analysis phantom used therein, and breast phantom - Google Patents

Description

The present invention relates to a method for manufacturing a phantom that reproduces the content of mammary gland tissue in a breast, an analytical phantom used in the method, and a breast phantom manufactured by the method.

X-rays have been used for a long time as a method for observing the state of a living body's organs from the outside. Furthermore, in the treatment of cancer, a method of destroying cancer cells by irradiating them with X-rays has been used for a long time. If a living body is irradiated with an excessive amount of X-rays, the living body will be damaged, so it is necessary to optimize the amount of X-rays to be irradiated.

Patent Document 1 describes the use of a model for the purpose of measuring the distribution of X-ray doses and calibrating an apparatus that irradiates X-rays. In the invention of Patent Document 1, a model is constructed by combining a plurality of cube-shaped blocks called solid water. A radiochromatographic film is placed inside the model for the purpose of measuring the X-ray dose. X-rays are irradiated from a radiation generator inserted inside the model, and the X-ray dose is measured using radiochromatography film. Based on the results, the X-ray dose distribution is determined. Patent Document 1 describes, as another example, a phantom in which the head is divided into a plurality of sections and subsections. Radiochromatographic film is placed between each section.

Patent Document 2 describes that a phantom imitating a lung is constructed from a plurality of cubic blocks. The phantom disclosed in Patent Document 2 is said to be constructed from a plurality of blocks having different moduli of elasticity in order to be compatible with dynamic image capturing using 4D-CT. This is said to make it possible to construct phantoms that mimic organs with different tissue densities, such as the lungs. For example, when the lungs inhale, the lower part of the lung expands more than the upper part. When you exhale, the lower part of your lungs contracts more than the upper part. In order to reproduce this difference in contraction rate, the lower part of the lung is thought to be constructed from a block with high elasticity, such as a sponge.

Special Publication No. 10-506298 U.S. Patent No. 9,778,374

Mammography has become popular as a method for examining the condition of the breasts. When performing mammography, the breast is compressed from above and below and stretched thin before being exposed to X-rays. Another technique used is to combine mammography with a technique called tomosynthesis, which takes cross-sectional images of the breast and observes the state of the breast by dividing it into multiple cross-sectional planes. Since these methods involve irradiation with X-rays, exposure to radiation is unavoidable. It is said that mammary gland tissue has a higher risk of developing cancer due to exposure to X-rays than breast skin. There are concerns that conducting tests with high doses of radiation or conducting tests frequently may increase the risk of developing breast cancer. Therefore, evaluation of X-ray exposure dose is important from a safety perspective.

Unlike the head example shown in Patent Document 1 or the lung example shown in Patent Document 2, in the case of breasts, the content of mammary gland tissue, fat content, or breast thickness varies depending on the person. different. Furthermore, the content of mammary glands increases or decreases depending on age even in the same person. Therefore, unlike the phantom described in Patent Document 1 or Patent Document 2, when optimizing the irradiation X-ray dose, the content of mammary gland tissue or breast thickness of each subject is taken into account. There is a need.

Currently, the exposure dose when performing mammography is evaluated by the average mammary gland dose according to the EUREF protocol of the European Organization for Standardization. The average mammary gland dose is the average dose absorbed by all mammary gland tissues, excluding the absorption of X-rays by subcutaneous fat. The average mammary gland volume is determined by the following formula.
[Formula 1]
D=K・g・s・c
However, D is the average mammary gland dose (mGy), K is the incident air kerma, g is the coefficient corresponding to 50% mammary gland content, and s is the target and filter with Mo-Mo as 1.0. c is a coefficient for correcting a mammary gland content rate different from the mammary gland content rate of 50%.

The average mammary gland dose calculated using the above formula is an estimated value based on a simulation using parameters such as breast thickness after compression, breast composition, X-ray tube voltage, half-value layer, etc. It is difficult to say that this reflects the variation in the dose absorbed into the mammary gland tissue due to factors such as the location of the mammary gland tissue, the distribution of the mammary gland tissue, subcutaneous fat, and age.

A rectangular plate-shaped PMMA phantom is known as a breast phantom, but the amount of radiation absorbed into the breast tissue depends on the position of the breast tissue, distribution of the breast tissue, subcutaneous fat, age, etc. of each subject. Variability is not taken into account.

An object of the present invention is to provide a breast phantom that reproduces the position and distribution of mammary gland tissue and the content of mammary gland tissue of an individual subject, a method for manufacturing the same, and an analytical phantom used for manufacturing the same.

The first step is to irradiate the compressed breast with X-rays to distinguish between mammary gland tissue and breast fat and take images. A second step of determining an average content rate of mammary gland tissue for each region, and a third step of arranging a plurality of block-shaped phantoms based on the average content rate and the position of the divided area. In the manufacturing method, the plurality of block-shaped phantoms reproduce the content rate of mammary gland tissue, and the plurality of block-shaped phantoms include multiple types with different content rates of mammary gland tissue to reproduce, The above problem is solved by a method for manufacturing a breast phantom in which block-shaped phantoms that reproduce the content of mammary gland tissue that matches or approximates the average content of breast tissue are arranged based on the position of the region. In this method, mammary gland tissue is divided into a plurality of regions having a predetermined size, and the average content of mammary gland tissue is determined for each region. Because the average is divided into smaller regions rather than the entire breast, it is possible to construct a breast phantom that takes into account the position, distribution, and mammary gland content of each subject's mammary gland tissue. In images of mammary gland tissue, the average content is shown for each sectioned area, not for each pixel. Therefore, it is easy to arrange a block-shaped phantom that reproduces the content rate of mammary gland tissue corresponding to the average content rate.

A method for analysis in which a plurality of block-shaped phantoms with a known mammary gland tissue content and different mammary gland tissue contents are arranged in a stepped manner, and have parts with different thicknesses and mammary gland tissue contents. Phantom solves the above problems. It becomes possible to analyze the correlation between the content of mammary gland tissue, the thickness of the breast, and the amount of X-rays that have passed through the analysis phantom and entered the detection unit using one phantom. Since there is no need to use multiple phantoms during analysis, analysis can be made more efficient.

A breast phantom is a breast phantom that is composed of a plurality of block-shaped phantoms. Breast phantoms containing different types solve the above problems. The breast phantom is composed of multiple types of block-shaped phantoms with different breast tissue contents, so the type and arrangement of the block-shaped phantoms used can be changed to match the breasts of each subject. It is possible to construct a phantom that reproduces the location, distribution, and content of mammary gland tissue of a subject.

In the breast phantom or the manufacturing method described above, it is preferable that an X-ray dose measuring means be disposed inside the breast phantom. For example, the X-ray dose measuring means is a plurality of dosimetry films, and in the breast phantom, the dosimetry films can be placed adjacent to the block-shaped phantom at the position where the exposure dose is to be measured. .

In the above manufacturing method, it is preferable that the shape of the divided region and the shape of the block-shaped phantom from an upper viewpoint be the same. Thereby, the block-shaped phantoms can be more easily arranged based on the average content of mammary gland tissue in the partitioned areas.

According to the present invention, there is provided a breast phantom that reproduces the breast thickness, the position and distribution of mammary gland tissue, and the content of mammary gland tissue of an individual subject, a method for manufacturing the same, and an analytical phantom used for manufacturing the same. be able to.

FIG. 2 is a diagram schematically showing an image of a breast including mammary gland tissue and fat. FIG. 1 is a schematic diagram illustrating the configuration of a mammography apparatus. It is a perspective view showing an example of a breast phantom. FIG. 4 is a plan view showing the breast phantom of FIG. 3 layer by layer. FIG. 4 is a perspective view showing an exploded state of the portion in which the mammary gland content is indicated in FIGS. 3 and 4. FIG. FIG. 1 is a schematic diagram illustrating the configuration of a mammography apparatus compatible with tomosynthesis. FIG. 3 is a diagram showing the distribution of X-ray doses absorbed by each block-shaped phantom when mammography is performed using a breast phantom. FIG. 2 is a diagram summarizing the X-ray doses incident on each GAFCHROMIC film placed in a breast phantom. It is a graph showing the correlation between the X-ray dose incident on the GAFCHROMIC film and the depth at which the GAFCHROMIC film is placed. It is a figure showing another example of composition of a breast phantom. It is a perspective view showing another example of composition of a breast phantom. This is an image obtained by irradiating a compressed breast with X-rays to distinguish between mammary gland tissue and breast fat. 13 is a diagram summarizing the averaged pixel values for each partitioned area after analyzing the image in FIG. 12. FIG. 13 is a diagram showing a state in which the image in FIG. 12 is divided into regions. FIG. It is a graph showing the correlation between pixel value and X-ray dose. 14 is a diagram summarizing the values obtained by converting the pixel values in FIG. 13 into the amount of X-rays transmitted through the breast using the correlation between the pixel values and the amount of X-rays. FIG. FIG. 2 is a perspective view of a block-shaped phantom used to construct an analytical phantom. FIG. 2 is an exploded perspective view of an analysis phantom. FIG. 2 is a perspective view showing how to use the analysis phantom. This is a table summarizing the relationship between the thickness of the analytical phantom (breast), the mammary gland content, and the X-ray dose transmitted through the analytical phantom (breast) when the analytical phantom is irradiated with X-rays. It is a triaxial graph showing the correlation between the thickness of the analysis phantom (breast), the mammary gland content, and the X-ray dose transmitted through the analysis phantom (breast). It is a graph showing the correlation between the mammary gland content rate and the X-ray dose transmitted through the analytical phantom (breast) when the thickness of the analytical phantom (breast) is 40 mm. FIG. 17 is a diagram in which the X-ray dose transmitted through the breast in FIG. 16 is converted into mammary gland content rate using the correlation between the mammary gland content rate and the X-ray dose transmitted through the analysis phantom (breast). FIG. 24 is a diagram illustrating the mammary gland content rates summarized in FIG. 23 by replacing them with approximate values. This is a diagram based on the mammary gland content rates summarized in FIG. 23 and replaced with approximate values using a different standard from that in FIG. 24.

Embodiments of the present invention will be described below with reference to the drawings. The embodiment shown in the drawings is only one example of a breast phantom, an analytical phantom, or a method of manufacturing a breast phantom. The technical scope of the present invention is not limited to the examples shown in the figures.

FIG. 1 shows a schematic diagram of an image of a breast taken by mammography. Breast 9 has mammary gland tissue 91, fat 92 covering mammary gland tissue 91, and skin tissue 93. As shown in FIG. 12, due to the difference in X-ray transmittance, mammary gland tissue 91 appears white and fat 92 appears black.

Mammography images such as those shown in FIG. 1 or 12 are captured using, for example, an apparatus such as that shown in FIG. 2. The apparatus 8 in FIG. 2 includes an imaging table 81 containing an X-ray detection section, a compression plate 82, and an X-ray tube 83. When taking an image of the breast 9, the breast 9 is sandwiched between the imaging table 81 and the compression plate 82 to make the breast thinner. By making the breast thinner, mammary gland tissue 91 and lesions can be more easily confirmed, and the radiation dose can also be reduced. Since the detection unit is located below the breast, the image in FIG. 1 or 12 is an image of the breast 9 taken from below.

In the method of this embodiment, the breast phantom 1 shown in FIG. 3 or 4 is manufactured. The breast phantom 1 is composed of a plurality of block-shaped phantoms 11. The plurality of block-shaped phantoms 11 reproduce the mammary gland tissue content rate (hereinafter referred to as mammary gland content rate) for each mammary gland tissue content rate. For example, in Figure 3 or 4, the areas marked with numbers "79," "93," "78," or "89" have a mammary gland content of 79% (fat 21%) as shown in Figure 5. %), 93% (7% fat), 78% (22% fat), and 89% (11% fat) materials that reproduce breasts. For example, a block-shaped phantom with a mammary gland content of 79% is configured to have the same X-ray transmittance as a breast with a mammary gland content of 79%. In other words, a block-shaped phantom with a mammary gland content of 79% is configured to be equivalent to a breast with a mammary gland content of 79%.

The higher the atomic number or density of a substance, the greater the degree to which it blocks X-rays. Furthermore, the greater the thickness, the greater the degree of blocking of X-rays. By selectively using materials with different thicknesses, atomic numbers, or densities, it is possible to manufacture a block-shaped phantom equivalent to a breast having any mammary gland content. For example, it is also possible to manufacture block-shaped phantoms that reproduce different mammary gland tissue contents in 1% increments.

As shown in FIG. 3 or 4, the breast phantom 1 is constructed to include a plurality of types of block-shaped phantoms 11 having different contents of breast tissue to be reproduced. By changing the arrangement of multiple types of block-shaped phantoms 11 with different mammary gland contents based on the distribution and position of the mammary gland tissue of the subject, it is possible to precisely reproduce the distribution and position of the mammary gland tissue of the subject's breast. . Additionally, the thickness of stacking the block-shaped phantoms 11, the number of block-shaped phantoms 11 used, the pattern in which the block-shaped phantoms 11 are arranged, and the thickness of the block-shaped phantoms 11 can be changed. This makes it possible to precisely reproduce the thickness, width, size, or shape of the subject's breasts. In the example of FIG. 3, the breast phantom 1 is constructed in a substantially triangular prism shape that imitates the shape of a compressed breast.

The breast phantom 1 shown in FIG. 3 or 4 is composed of multiple layers. As shown in FIG. 3 or 4, the breast phantom 1 includes a first layer 111 formed by laying a plurality of block-shaped phantoms 11, and a second layer 111 formed by laying a plurality of block-shaped phantoms 11. The third layer 113 includes a layer 112 and a third layer 113 formed by laying a plurality of block-shaped phantoms 11. In the example of the breast phantom 1, as shown in FIGS. 3 to 5, where some of the numerical values of mammary gland content are filled in, the subject is divided into multiple layers such as a first layer 111, a second layer 112, or a third layer 113. It reproduces the mammary gland content of Breast phantoms that reproduce mammary gland content in multiple layers are manufactured by creating blueprints for each layer using mammography using tomosynthesis (tomography) shown in Figure 6, as described later. can do.

As shown in FIG. 5, inside the breast phantom 1, a GAFCHROMIC film 12 is arranged as an X-ray dose measuring means. The GAFCHROMIC film 12 is placed adjacent to the block-shaped phantom 11 located at the position where the exposure dose is to be measured. In the example of FIG. 5, in the vertical direction, on the block-shaped phantom 11 constituting the first layer, between the block-shaped phantom 11 constituting the first layer 111 and the block-shaped phantom 11 constituting the second layer 112. , a GAFCHROMIC film 12 is disposed between the block-shaped phantom 11 forming the second layer 112 and the block-shaped phantom 11 forming the third layer 113, and below the block-shaped phantom 11 forming the third layer. The block-shaped phantom has a thickness of 10 mm. Therefore, the GAFCHROMIC film 12 is located at the "surface", "10 mm depth", "20 mm depth", and "30 mm depth" of the breast phantom. In the horizontal direction, the GAFCHROMIC film 12 may be arranged between each block-like phantom.

A method of using the breast phantom 1 will be explained. The above breast phantom 1 is installed on the imaging table 81, 81a of the mammography apparatus 8, 8a in FIG. 2 or 6, and the breast phantom 1 is sandwiched between the imaging table 81, 81a and the compression plate 82, 82a. . In this state, the breast phantom 1 is irradiated with X-rays from the X-ray tubes 83 and 83a. The phantom 1 is removed from the apparatus 8 and the dose incident on each GAFCHROMIC film 12 is analyzed. To analyze the dose incident on a GAFCHROMIC film, for example, the degree of exposure of the film is converted into pixel values, and the dose is read using a calibration curve. The calibration curve is created by irradiating the GAFCHROMIC film with a known dose of X-rays multiple times at different doses to determine the correlation between pixel values and dose. Create a fitting function or graph from the correlation and use it as a calibration curve. To determine the degree to which the GAFCHROMIC film has been exposed to light, scan the GAFCHROMIC film. The image is then analyzed for the pixel values of the pixels in the exposed area. If the image is RGB, as will be described later, any one of the red element, green element, and blue element may be extracted and used as a pixel value. For example, the red component can be extracted and used as a pixel value.

Analysis of the X-ray dose absorbed by the breast phantom is performed as follows. First, as shown in FIG. 8, the X-ray dose incident on a total of four GAFCHROMIC films 12 at the surface, at a depth of 10 mm, at a depth of 20 mm, and at a depth of 30 mm is determined by information on X, Y, and Z coordinates. Record in association with. The Z axis indicates the position of the layers of the breast phantom. X indicates the lateral position of any layer of the breast phantom. Y indicates the position of an arbitrary layer of the breast phantom in the longitudinal direction.

Next, for each block-shaped phantom constituting the breast phantom, the correlation between the X-ray dose absorbed by the GAFCHROMIC film and the depth at which the GAFCHROMIC film is placed between two GAFCHROMIC films adjacent in the vertical direction Seeking a relationship. The correlation can be determined by plotting the relationship between the X-ray dose and the depth at which the GAFCHROMIC film is placed on a graph and drawing an approximate curve, as shown in FIG. Alternatively, a function (approximate curve) that fits the correlation with the depth may be obtained by curve fitting.

An example of a method for analyzing the X-ray dose absorbed by a breast phantom is shown below. In this example, in FIG. 8, the X-ray dose absorbed by the block-shaped phantom with X=7, Y=B, and Z=1 to 3 is determined. In the block-shaped phantom with X=7 and Y=B, there is a correlation shown in Figure 9 between the X-ray dose (mGy) incident on the GAFCHROMIC film and the depth (mm) at which the GAFCHROMIC film is placed. To establish. In other words, in the depth range of 0 to 10 mm, the correlation of Y = 16.12e ^-0.77x is established, and in the depth of 10 to 20 mm, the correlation of Y = 20.476e ^-0.101x is established, and the depth In the range of 30 to 40 mm, the correlation of Y=11.317e ^-0.071x holds true. As shown in FIG. 9, the area in the depth range of 0 to 10 mm, that is, the integral value corresponds to the X-ray dose absorbed by the block-shaped phantom with X=7, Y=B, and Z=1. Then, the area in the depth range of 10 to 20 mm, that is, the integral value corresponds to the X-ray dose absorbed by the block-shaped phantom with X=7, Y=B, and Z=2. The area in the depth range of 20 to 30 mm, that is, the integral value corresponds to the X-ray dose absorbed by the block-shaped phantom with X=7, Y=B, and Z=3. In the example of FIG. 9, the integrated value from depth 0 to 10 mm, that is, the X-ray dose absorbed by the block-shaped phantom in the first layer is 11.04 mGy. The integral value at a depth of 10 to 20 mm, that is, the X-ray dose absorbed by the block-shaped phantom in the second layer is 4.68 mGy. The integrated value at a depth of 20 to 30 mm, that is, the X-ray dose absorbed by the block-shaped phantom in the third layer is 1.94 mGy. In this way, the thickness range of the block-like phantoms that make up the breast phantom is specified, and the correlation between the thickness of the breast phantom and the X-ray dose is used to calculate the amount of radiation absorbed by the block-like phantoms existing within the thickness range. By integrating the X-ray doses, the X-ray dose absorbed by the block-shaped phantom can be determined. The reason why the relationship between the X-ray dose and the thickness of the breast phantom is analyzed for each layer is because the breast phantom used in this example has a different mammary gland content in each layer.

The analysis of the X-ray dose absorbed by the breast phantom was performed using a RAW image generated based on information on the X-ray dose incident on the detection unit of the mammography device, instead of the image obtained by scanning the GAFCHROMIC film described above. It's okay. A RAW image is an image based on unprocessed data. By using tomography, it is possible to evaluate the amount of X-ray absorption by dividing the breast phantom into multiple layers by tomographically imaging the breast phantom.

In the above manner, the absorbed dose of X-rays can be evaluated for each breast position, as shown in FIG. Each cell in FIG. 7 corresponds to a plurality of block-shaped phantoms 11 constituting the breast phantom 1, and each cell reflects the position of each block-shaped phantom. In each cell in FIG. 7, the X-ray absorbed dose (mGy) of the block-shaped phantom 11 at that position is written. "First stage", "second stage", and "third stage" in FIG. 7 correspond to the first layer 111, the second layer 112, or the third layer 113, respectively. If the X-ray dose of the mammography device is calibrated for each subject based on this X-ray absorbed dose distribution data, unnecessary exposure can be reduced and mammography can be performed under conditions that allow diagnosis of breast conditions. It becomes possible.

In the breast phantom 1 described above, the breast was divided into a plurality of layers based on tomosynthesis data, and each layer was divided into a plurality of divided regions to reproduce the mammary gland content. When mammary gland tissue is imaged by normal mammography rather than tomosynthesis tomography, data on the mammary gland content of a single layer is obtained. In this case, as shown in FIG. 10, a plurality of block-shaped phantoms 11 may be arranged so that the mammary gland content rates of the first layer 111a, the second layer 112a, and the third layer 113a are the same. In the example of FIG. 10, the first layer 111a, second layer 112a, and third layer 113a of the breast phantom 11a are constructed using block-shaped phantoms 11 with a mammary gland content of 11%. In the section next to that area, a first layer 111a, a second layer 112a, and a third layer 113a are constructed of block-shaped phantoms 11 with a mammary gland content of 47%.

When data on the mammary gland content of a single layer is obtained by ordinary mammography, as shown in FIG. may be arranged to form a single-layer breast phantom 11b.

The detection means may be anything that can measure the dose of X-rays, and for example, a dose measurement film such as a GAFCHROMIC film, a small X-ray detector, etc. can be used. It is preferable that the detection means be placed adjacent to the block-shaped phantom located at the position where the exposure dose is to be detected.

As the block-shaped phantom for constructing the breast phantom, one having a shape such as a cube, a triangular prism, or a quadrangular prism can be used.

Next, a method for manufacturing the above-mentioned breast phantom will be explained. In order to simplify the explanation, below, a method for capturing a single-layer breast image by normal mammography without using tomography and constructing a breast phantom based on the image will be described.

First, a first step is performed in which the compressed breast is irradiated with X-rays to distinguish and image breast tissue and breast fat. The first step can be performed using a known mammography device. FIG. 12 shows a captured image containing breast tissue 91 and fat 92. In FIG. 12, a portion photographed in a color close to black is fat 92, and a portion photographed in a color close to white is mammary gland tissue 91.

Following the first step, a second step is performed. In the second step, the mammary gland tissue 91 and fat 92 obtained in the first step are used to divide the breast image into a plurality of regions having a predetermined size. In the example of FIG. 12, one section is set as a rectangular area, and the image of FIG. 12 is partitioned into a total of 260 areas without gaps, as shown in FIGS. 12 and 14. Then, the average content of mammary gland tissue is determined for each region. The number of partitioned areas is not particularly limited and can be changed as appropriate.

To find the average content of mammary gland tissue for each region, use image analysis to find the number of pixels included in one divided region, add up the pixel values of the pixels included in that one region, and calculate the total value. Find the average pixel value by dividing by the number of pixels. The average value of the determined pixel values is recorded in association with the coordinates (position) of each region, as shown in FIG. FIG. 13 shows the average value of pixel values recorded in cells corresponding to each area. Note that in this method, a breast phantom is constructed based on a single-layer breast image. Therefore, the coordinates of each area described above are two values, the X axis and the Y axis, and these two values and the pixel value are recorded in association with each other. When constructing a phantom that can evaluate the three-dimensional distribution of X-ray absorption in the breast by taking multiple tomographic images of the breast using tomography, it is necessary to These three values and the pixel value are recorded in association with each other. For example, the pixel value of X=6, Y=10, Z=1 (first layer) in FIG. 13 is 1138.992. These numerical values are recorded as a table or numerical values as shown in FIG.

When calculating the pixel value of the above pixel, if it is a grayscale image, sum the pixel values of each pixel included in the divided area, and divide the total value by the number of pixels included in the divided area. It can be found by For RGB, extract one of the red, green, and blue elements, find the total value of the extracted elements, and divide the total value by the number of pixels included in the corresponding area. It may be averaged. First, for each pixel, each pixel value (R, G, B) is summed, divided by 3, and the average value of RGB of each pixel is determined by the NTSC weighted average method. Next, the average RGB values of pixels included in the applicable area by the NTSC weighted average method are summed, and the total value is divided by the number of pixels included in the applicable area to obtain the average value of the pixel values in the applicable area. You can also use it as

Next, a correlation between pixel values and incident dose is determined. A known radiation dose of X-rays is made to enter the film, imaging plate, or flat panel that constitutes the detection part of the mammography device multiple times, and pixel values are recorded in relation to the incident radiation dose, and the pixel values and the incident Determine dose correlation. For example, as shown in FIG. 15, the detection unit is irradiated with X-rays of known dose (mGy) multiple times, and the pixel values at that time are recorded. A function (approximate curve) that fits this data is found by curve fitting. In the example of FIG. 15, the following relationship holds true between the pixel value (Y) and the dose (X) incident on the detection unit.
[Formula 2]
Y=-355.8ln(x)+982.48

Using Equation 2 above or a graph showing the correlation between pixel values and X-ray dose such as the constituent curve in Figure 15, the average value of the pixel values in each region shown in Figure 13 can be calculated using the Convert to line dose. As shown in FIG. 16, the converted X-ray dose (mGy) transmitted through the breast is recorded in association with the coordinates of each region. Each cell in FIG. 16 corresponds to each partitioned area. The number recorded in each cell is the X-ray dose (mGy) transmitted through the breast. In the method of FIG. 16, a breast phantom is constructed based on a single layer breast image. Therefore, the coordinates of each region described above are two values, the X-axis and the Y-axis, and these two coordinate values and the X-ray dose transmitted through the breast are recorded in association with each other. When constructing a phantom that can evaluate the three-dimensional distribution of X-ray absorption in the breast by taking multiple tomographic images of the breast using tomography, the coordinates are the X-axis, Y-axis, and Z-axis. These three values are recorded in association with the X-ray dose transmitted through the breast.

Next, we constructed the analysis phantom shown in FIGS. 18 and 19, and calculated the amount of X-rays that passed through the analysis phantom and entered the detection section, the thickness of the analysis phantom at the part irradiated with X-rays, and the Find the correlation with the mammary gland tissue content of the block-shaped phantom irradiated with radiation. As will be described in detail later, this correlation is used to convert the X-ray dose transmitted through the breast determined in FIG. 16 into the average content of mammary gland tissue.

First, the analysis phantom 4 shown in FIGS. 18 and 19 is constructed using the plurality of block-shaped phantoms 11 shown in FIG. 17. The block-shaped phantom 11 can be the same as that used when constructing the breast phantom 1 described above.

As shown in FIGS. 18 and 19, the analysis phantom 4 includes a plurality of block-shaped phantoms 11 having known mammary gland tissue content rates and different mammary gland tissue content rates arranged with steps 41. It has portions with different thicknesses and mammary gland tissue contents. Specifically, the analysis phantom 4 is composed of five types of block-shaped phantoms 11 with mammary gland content rates of 0%, 25%, 50%, 75%, and 100%. The analysis phantom 4 has a plurality of steps, that is, portions with different thicknesses. The plurality of steps 41 are constructed by stacking a plurality of layers having different areas including a first layer, a second layer, a third layer, a fourth layer, and a fifth layer.

In the analysis phantom 4 shown in FIG. 19, the block-shaped phantoms 11 stacked vertically have the same mammary gland content and are arranged with steps. For example, there is a part where block-like phantoms with a mammary gland content of 0% are stacked in five stages, and there is also a part where block-like phantoms with a mammary gland content of 0% are stacked in four stages. Furthermore, in the analysis phantom 4 shown in FIG. 19, the block-shaped phantoms arranged in the horizontal direction have different mammary gland content rates. For example, the first stage 411 is composed of a plurality of block-shaped phantoms having different mammary gland content rates of 0%, 25%, 50%, 75%, and 100%. The second stage 412, the third stage 413, the fourth stage 414, and the fifth stage 415 are also a plurality of two-layer block-shaped phantoms with different mammary gland contents of 0%, 25%, 50%, 75%, and 100%. It consists of

For example, as shown in FIG. 19, if X-rays are irradiated onto the fifth layer 415 where five block-shaped phantoms with a mammary gland content of 0% are stacked, and the amount of transmitted X-rays is measured by the detection unit. , the transmitted X-ray dose can be determined when the mammary gland content is 0% and the thickness is 5 steps. Then, if X-rays are irradiated to the fourth layer 414 where the block-shaped phantoms with a mammary gland content of 0% are stacked in four stages and the amount of transmitted X-rays is measured by the detection unit, the mammary gland content is 0%, the thickness is The transmitted X-ray dose for four stages can be determined. Then, if the block-shaped phantom in the first stage 411 with a mammary gland content of 25% is irradiated with X-rays and the transmitted X-ray dose is measured by the detection unit, the transmitted X-ray dose can be determined. Then, if the block-shaped phantom in the first stage 411 with a mammary gland content of 50% is irradiated with X-rays and the transmitted X-ray dose is measured by the detection unit, the transmitted X-ray dose can be determined.

As mentioned above, according to the analytical phantom 4 shown in FIGS. 18 and 19, if the analytical phantom 4 is mounted on a mammography apparatus, analysis can be performed in various combinations by appropriately changing the part irradiated with X-rays. Data on the thickness of the phantom (thickness of the breast), mammary gland content, and transmitted X-ray dose can be obtained. Since data related to various combinations can be obtained with one analysis phantom 4, it is possible to eliminate the complexity of preparing multiple analysis phantoms or transferring multiple analysis phantoms to a mammography apparatus. can.

Next, using the analysis phantom 4, calculate the amount of X-rays that passed through the analysis phantom 4 and entered the detection section, the thickness of the analysis phantom at the part irradiated with the X-rays, and the block irradiated with the X-rays. A method for determining the correlation with the mammary gland content of the phantom will be explained.

First, the analysis phantom 4 is set in the mammography apparatus, different parts of the analysis phantom 4 are irradiated with X-rays, and the X-ray dose (mGy) transmitted through the analysis phantom 4 and the part irradiated with The thickness (mm) and the known mammary gland content (%) of the block-shaped phantom 11 are recorded in association with each other, as shown in FIG. In the example of FIG. 20, a table is used to achieve association. In the example of FIG. 20, the analysis phantom 4 is irradiated with X-rays three times in total, and the average dose of the transmitted X-rays is determined. Note that the measurement conditions were a room temperature of 18.4° C., an atmospheric pressure of 993 hPa, an X-ray tube: molybdenum, a filter: molybdenum, and a tube voltage: 30 kV.

Based on the data of the X-ray dose (mGy) transmitted through the analytical phantom 4, the thickness (mm) of the part irradiated with X-rays, and the known mammary gland content (%) of the block-shaped phantom 11, the analytical phantom is The correlation between the amount of X-rays transmitted through the phantom and incident on the detection unit, the thickness of the analysis phantom at the portion irradiated with X-rays, and the content of mammary gland tissue in the block-shaped phantom irradiated with X-rays is determined. This correlation is, for example, between the X-ray dose (mGy) transmitted through the analytical phantom 4, the thickness (mm) of the portion irradiated with X-rays, and the known mammary gland content rate (%) of the block-shaped phantom 11. , and can be analyzed by plotting it on a graph and creating a three-axis graph. FIG. 21 shows an example of a three-axis graph. In the three-axis graph in Figure 21, the X-axis is the thickness of the analytical phantom (mm), the Y-axis is the mammary gland content (%) of the block-shaped phantom, and the Z-axis is the amount of X-rays transmitted through the analytical phantom. dose (mGy).

The correlation between the amount of X-rays transmitted through the analytical phantom and incident on the detection part, the thickness of the analytical phantom in the area irradiated with X-rays, and the content of mammary gland tissue in the block-shaped phantom irradiated with X-rays is as follows. If the thickness of the analytical phantom is taken as a constant, it can also be analyzed as a correlation between the amount of X-rays transmitted through the analytical phantom and incident on the detection unit and the content of mammary gland tissue in the block-shaped phantom irradiated with X-rays. I can do it. Specifically, a function that fits the correlation between data on the X-ray dose transmitted through the analysis phantom and incident on the detection unit at a specific thickness and the content of mammary gland tissue in the block-shaped phantom irradiated with X-rays. (approximate curve) can be obtained by curve fitting or analyzed as a graph of the approximate curve. Regarding the thickness of the analysis phantom, the thickness of the compressed breast is measured when performing mammography of the subject's breast in step 1 above. The thickness of the subject's breast measured at this time may be used as the thickness of the analysis phantom.

Figure 22 is a graph showing the correlation between the X-ray dose (Y) transmitted through the analytical phantom and the mammary gland content (X) when the previously measured breast thickness of the subject was 40 mm. be. The following relationship holds between X and Y.
[Formula 3]
Y=0.2444e ^-0.013X

The result of FIG. 16 described above, that is, the data summarizing the X-ray dose transmitted through the breast by irradiating the subject's breast with each divided region, and the three-axis graph of FIG. 21, or the above formula Using the function shown in 3 or the graph of FIG. 22, the X-ray dose transmitted through the breast is converted into mammary gland content. Note that when reading the mammary gland content rate from the three-axis graph of FIG. 21, the actual breast thickness of the subject measured in advance in step 1 is used, as described above. In this example, the thickness of the subject's breast is 40 mm, as described above.

The results of converting the X-ray dose transmitted through the breast into mammary gland content are recorded for each coordinate of the divided area, as shown in FIG. The coordinates of each region described above are two values, an X axis and a Y axis, and these two values and mammary gland content are recorded in association with each other. Note that the mammary gland content rate recorded in FIG. 23 is obtained by averaging pixel values from breast images, and therefore corresponds to the average content rate of mammary gland tissue. When constructing a phantom that can evaluate the three-dimensional distribution of X-ray absorption in the breast by taking multiple tomographic images of the breast using tomography, it is necessary to These three values and mammary gland content are recorded in association with each other.

The numerical value written in each cell in FIG. 23 is the mammary gland content rate (%) calculated from the breast image of the subject. If the block-shaped phantom 11 equivalent to the mammary gland content of each cell (each region) and the X-ray dose detection means are arranged using FIG. 23 as a design drawing, the distribution of the subject's breast tissue as shown in FIG. 10, etc. Breast phantom 1 that reproduces can be constructed. Note that the position where the block-shaped phantom is to be placed can be easily determined by reading the coordinates in the design drawing of FIG. 23. For example, it can be read from FIG. 23 that a block-shaped phantom equivalent to a mammary gland content of 47% should be placed at the position of X=11 and Y=4.

Note that the thickness and width of the compressed breast, which were measured in advance in step 1, are recorded. When constructing a breast phantom as shown in FIG. 10 using the block-shaped phantom 11, not only the thickness of the breast but also the width of the breast is reproduced. In the method of this embodiment, the shape of the block-shaped phantom when viewed from above matches the shape of the partitioned region. This makes it easier to arrange block-shaped phantoms based on the design drawings. The dimensions of the divided region on the image and the dimensions of the block-shaped phantom may be different as long as the shapes are similar.

FIG. 23 shows the distribution of mammary gland content consisting of a single layer determined by conventional mammography. For example, when tomosynthesis is used to image a breast divided into three layers, a design diagram of the second layer and a design diagram of the third layer corresponding to FIG. 23 are also obtained. It is also possible to construct a phantom that three-dimensionally reproduces the mammary gland content of a subject using three layers of blueprints. Note that in tomosynthesis, as shown in FIG. 6, a device 8a that can displace the X-ray tube 83a is used. The breast 9 sandwiched between the compression plate 83a and the imaging table 81a is irradiated with X-rays while changing the position of the X-ray tube to focus on a desired cutting layer. A tomographic image can be obtained.

In the example of FIG. 23, a breast phantom is constructed using a plurality of block-shaped phantoms 11 having a mammary gland content rate that matches the mammary gland content rate described in the design drawing of FIG. 23. A design diagram as shown in FIG. 24 or 25 may be created for the purpose of constructing a breast phantom using a block-shaped phantom that approximates the mammary gland content of the subject. In FIG. 24 or 25, an approximate value of the mammary gland content rate is calculated based on the mammary gland content rate in FIG. 23 determined based on the image of the mammary gland tissue of the breast, and the approximate value is recorded for each sectioned area. It is something. In each cell in FIG. 24 or 25, an approximate value (%) of the mammary gland content rate for each sectioned area is written.

In the example of FIG. 24, the approximate value is obtained as follows. When the mammary gland content rate recorded in the divided area of FIG. 23, that is, the cell of FIG. 23, falls within the range of 0% or more and less than 12.5%, the approximate value of the mammary gland content rate is set to 0%; When the mammary gland content rate recorded in the cell of Figure 23 falls within the range of 12.5% or more and less than 37.5%, the approximate value of the mammary gland content rate is set to 25%; the mammary gland content rate recorded in the cell of Figure 23 When the mammary gland content is in the range of 37.5% or more and less than 62.5%, the approximate value of the mammary gland content is set to 50%; When the mammary gland content falls within the range of less than 87.5%, the approximate value of the mammary gland content is set to 75%; when the mammary gland content recorded in the cell of Figure 23 falls within the range of 87.5% or more, the mammary gland content Let the approximate value of be 100%.

In the example of FIG. 25, the approximate value was obtained as follows. When the mammary gland content rate recorded in the divided area of FIG. 23, that is, the cell of FIG. 23, falls within the range of 0% or more and less than 6.25%, the approximate value of the mammary gland content rate is set to 0%; When the mammary gland content rate recorded in the cell of Figure 23 falls within the range of 6.25% or more and less than 18.75%, the mammary gland content rate is approximated to 12.5%; When the content rate falls within the range of 18.75% or more and less than 31.25%, the approximate value of the mammary gland content rate is set to 25%; When the mammary gland content falls within the range of less than 43.75%, the approximate value of the mammary gland content is 37.5%; the mammary gland content recorded in the cell of Figure 23 is in the range of 43.75% or more and less than 56.25%. If the mammary gland content is within the range of 56.25% or more and less than 68.75%, the mammary gland content is approximated as 50%. The approximate value is set to 62.5%; when the mammary gland content rate recorded in the cell of Figure 23 falls within the range of 68.75% or more and less than 81.25%, the approximate value of the mammary gland content rate is set to 75%. ; When the mammary gland content rate recorded in the cell of Figure 23 falls within the range of 81.25% or more and less than 93.5%, the approximate value of the mammary gland content rate is set to 87.5%; When the recorded mammary gland content falls within the range of 93.75% or more, the approximate value of the mammary gland content is taken as 100%.

When constructing a breast phantom based on the design diagram of FIG. 24, a total of five types of block-shaped phantoms equivalent to a mammary gland content of 0%, 25%, 50%, 75%, or 100% are used. When constructing a breast phantom based on the blueprint in Figure 25, 0%, 12.5%, 25%, 37.5%, 50%, 62.5%, 75%, 87.5%, or A total of nine types of block-shaped phantoms equivalent to 100% mammary gland content are used. Compared to the case of using the design diagram in 1% increments as in the case of FIG. 23, the design diagram of FIG. 24 or 25 requires fewer block-shaped phantoms, so it takes more effort to construct the breast phantom. can be reduced.

When constructing a breast phantom based on the design diagram in FIG. 24, a plurality of block-shaped phantoms are stacked in the vertical direction, and the average value of the mammary gland content of the plurality of block-shaped phantoms stacked in the vertical direction. may coincide with the partitioned area shown in the design drawing of FIG. For example, in FIG. 24, the mammary gland content in the region of X=13 and Y=2 is 25%. When the block-shaped phantom is composed of four layers, four block-shaped phantoms are stacked. Then, one block-shaped phantom with a mammary gland content of 100% and a block-shaped phantom with a mammary gland content of 0% are used. In this case, the mammary gland content in the region of X=13 and Y=2 of the mammary gland phantom is 25% according to the following equation. In this way, the number of block-shaped phantoms to be used can be reduced to two, making it easier to construct breast phantoms and manage block-shaped phantoms. In this case, a 100% block-shaped phantom with a high mammary gland content is positioned at the center of the mammary phantom. In this example, the center is the second or third layer. In this way, when arranging block-shaped phantoms that reproduce approximate mammary gland content rates, the average value of the mammary gland content rates of the block-shaped phantoms in the thickness direction may be made to match the mammary gland content rates.
[Formula 4] (100%+0%+0%+0%)÷4=25%

In the example of FIG. 24, a breast phantom imitating a breast with a thickness of 40 mm is constructed. For this reason, the block-shaped phantom is made by laminating four layers of cube-shaped blocks measuring 10 mm x 10 mm x 10 mm. The number of laminated layers and the dimensions of the blocks can be changed as appropriate.

In the example shown in Fig. 25, three types of block-shaped phantoms with mammary gland content of 0%, 50%, and 100% are used, and the breast phantom is composed of four layers, thereby reducing the number of block-shaped phantoms used. It is possible to reduce the number to three.

In the above manufacturing method, the number of regions defining the breast is not particularly limited and can be determined as appropriate. The method for determining the average content of mammary gland tissue is not particularly limited. For example, based on a breast image obtained by mammography, the mammary gland content can be determined for each pixel of the image using publicly available software. , the obtained values may be averaged for each partitioned area.

In the above manufacturing method, the numerical range for arranging block-shaped phantoms that approximate the mammary gland content shown in the design drawings is not particularly limited, and may be determined as appropriate. A block-shaped phantom having a predetermined mammary gland content may be arranged based on the mammary gland content shown in the design drawing.

The number of stages of the breast phantom is not particularly limited and can be changed as appropriate. The shape of the breast phantom is also not particularly limited, and may be changed depending on the shape of the breast of the subject. The shape of the analysis phantom is not limited to the above-mentioned step-like shape, and may have different thicknesses and mammary gland tissue contents.

As the above-mentioned block-shaped phantom, a cubic one measuring 10 mm x 10 mm x 10 mm was used. The dimensions of the block-shaped phantom can be changed as appropriate.

In the above example, pixel values from a grayscale image were used for mammography breast image analysis. The number of tones of a grayscale image differs depending on the device, so it is not particularly limited. Pixel analysis may use RGB and CMYK color values.

1 Breast phantom 4 Analysis phantom 11 Block-shaped phantom 91 Breast tissue 92 Fat

Claims (4)

A first step of irradiating the compressed breast with X-rays to distinguish between mammary gland tissue and breast fat and take an image;
a second step of dividing the mammary gland tissue into a plurality of regions having a predetermined size and calculating the average content of the mammary gland tissue for each region based on the image of the mammary gland tissue;
A method for manufacturing a breast phantom, comprising a third step of arranging a plurality of block-shaped phantoms based on the average content rate and the position of the divided region,
Multiple block-shaped phantoms reproduce the content of mammary gland tissue,
The multiple block-shaped phantoms include multiple types with different content rates of reproduced mammary gland tissue.
A method for manufacturing a breast phantom, in which block-shaped phantoms that reproduce a content rate of mammary gland tissue that matches or approximates the average content rate of mammary gland tissue are arranged based on the position of the region,
The second step to determine the average content of mammary gland tissue is
The reproduced mammary gland tissue content is known, and by arranging multiple block-shaped phantoms with different mammary gland tissue contents in a stepped manner, parts with different thicknesses and mammary gland tissue contents can be created. irradiating the analysis phantom with X-rays, and determining the correlation between the X-ray dose that has passed through the analysis phantom and entered the detection unit, and the content of mammary gland tissue in the block-shaped phantom that has been irradiated with the X-rays. and,
and determining a correlation between the X-ray dose and the pixel value,
The image containing mammary gland tissue and fat captured in the first step is divided into a plurality of regions having a predetermined size, the average value of pixel values is determined for each region, and the average value of the pixel values is determined. Using the correlation between the X-ray dose and the pixel value, determine the X-ray dose transmitted through the breast for each sectioned area,
Using the correlation between the X-ray dose transmitted through the breast, the X-ray dose transmitted through the analysis phantom and incident on the detection unit, and the content of mammary gland tissue in the block-shaped phantom irradiated with X-rays, the A breast phantom manufacturing method that calculates the average content of mammary gland tissue in each region. 2. The method for manufacturing a breast phantom according to claim 1 , wherein an X-ray dose measuring means is arranged inside the breast phantom. The X-ray dose measurement means is a plurality of dose measurement films,
3. The method of manufacturing a breast phantom according to claim 2 , wherein in the breast phantom, the dose measurement film is placed adjacent to the block-shaped phantom at a position where the exposure dose is to be measured. 4. The method of manufacturing a phantom according to claim 1 , wherein the shape of the divided region and the shape of the block-shaped phantom when viewed from above match.

Images