Distribution calculation device, method, and program

Abstract

Provided are a distribution calculation device, method, and program whereby the distribution of a physical value in a sample can be calculated from radiation CT data even when there are no calibration data obtained from a calibration sample. The present invention comprises a discrimination unit 340 for using radiation CT data obtained by measuring an object sample formed from a material having a single composition to discriminate between a region of the object sample and a region on the periphery of the object sample in the radiation CT data, a conversion coefficient calculation unit 370 for calculating a conversion coefficient for converting the luminance value of the object sample in each pixel of the region of the object sample to a mass density, and a distribution calculation unit 380 for using the conversion coefficient to calculate the distribution of a physical value in the object sample from the luminance values of the radiation CT data. By thus calculating a conversion coefficient using luminance values of an object sample formed from a material having a single composition, the distribution of a physical value in the object sample can be calculated from radiation CT data without a separately prepared calibration sample.

Description

Technical Field

[0001] The present invention relates to a distribution calculation device, method, and program for calculating the distribution of physical property values in a sample.

Background Art

[0002] Conventionally, in the measurement of physical property values of biological tissues, a calibration line is created from the luminance values of CT images obtained from a plurality of calibration samples having different densities imitating bone, and a method for obtaining human bone density is known (see Patent Document 1). For example, in the method described in Patent Document 1, first, two or more types of calibration jigs composed of a single element or compound are used, each calibration jig is measured in a plurality of energy regions, and the absorption coefficient is obtained from the detected data. Then, a calibration coefficient is calculated from the absorption coefficient, and the electron density of the measurement target is calculated based on the absorption coefficient in each energy region of the measurement target and the calibration coefficient in each energy region.

[0003] Also in the industrial field, for example, formed products such as resins, powders, or polymer gels, even if formed from the same material, have physical property values that vary depending on the forming conditions and the forming location. Therefore, in order to examine the forming conditions and analyze the characteristic values, it is necessary to grasp the three-dimensional distribution of physical property values.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the above method, it is necessary to prepare a calibration sample for each measurement target, and it is not practical to adopt it in fields where various samples are measurement targets. Also, basically, it is difficult to prepare a calibration sample with exactly the same composition as the sample to be measured. Even if a calibration sample with a different density is prepared and a calibration curve is obtained, an error will occur with respect to the true density distribution of the measurement target. In particular, for formed products that are not raw materials, when there is no data on physical property values such as composition, it is impossible to prepare a calibration sample and measure its density.

[0006] The present invention has been made in view of such circumstances, and an object thereof is to provide a distribution calculation device, method, and program capable of calculating the distribution of physical property values in a sample from radiation CT data even without calibration data obtained from a calibration sample.

Means for Solving the Problems

[0007] (1) To achieve the above object, the distribution calculation device of the present invention uses radiation CT data obtained by measuring a target sample formed of a material having a single composition, and separates the region of the target sample and the region around the target sample in the radiation CT data. A separation unit, a conversion coefficient calculation unit that calculates a conversion coefficient for converting the luminance value of the target sample at each pixel in the region of the target sample into a mass density, and using the conversion coefficient, from the luminance value of the radiation CT data. And a distribution calculation unit for calculating the distribution of physical property values in the target sample.

[0008] (2) Further, the distribution calculation device of the present invention further includes an average mass density calculation unit that calculates the volume of the separated region of the target sample and calculates the average mass density of the target sample using the calculated volume and the weight value of the target sample. It is characterized by that.

[0009] (3) Further, in the distribution calculation device of the present invention, the conversion coefficient calculation unit calculates the conversion coefficient by dividing the difference between the average luminance value of the target sample and the average luminance value around the target sample by the difference between the average mass density of the target sample and the average mass density around the target sample. It is characterized by that.

[0010] (4) Further, the distribution calculation device of the present invention is characterized in that the radiation CT data obtained by measuring the target sample is data reconstructed from the projection image data obtained under the condition that the target sample fits within the region where the projection image data is acquired.

[0011] (5) Further, the distribution calculation device of the present invention is characterized by including a reconstruction unit that can select a reconstruction calculation method for supplementing the condition for the projection image data obtained without satisfying the condition that the target sample fits within the region where the projection image data is acquired.

[0012] (6) Further, the distribution calculation device of the present invention is characterized in that the average mass density of the target sample is calculated using the volume calculated using the radiation CT data obtained by measuring a separate reference sample derived from the original sample of the target sample and the weight value of the reference sample.

[0013] (7) Further, the distribution calculation device of the present invention is characterized in that the distribution calculation unit calculates the distribution of any one of the filling rate, porosity, and crystallinity of the target sample from the distribution of the mass density based on a conversion value correlated with the mass density.

[0014] (8) Further, the distribution calculation device of the present invention is characterized in that the distribution calculation unit calculates the change amount of the physical property value of the target sample in each pixel from the distribution of the physical property value calculated from the CT data of the target sample before and after the environmental change.

[0015] (9) Further, the distribution calculation method of the present invention includes the steps of: using radiation CT data obtained by measuring a target sample formed of a material having a single composition, separating the region of the target sample and the region around the target sample in the radiation CT data; calculating a conversion coefficient for converting the luminance value of the target sample in each pixel of the region of the target sample into a mass density; and calculating the distribution of the physical property value in the target sample from the luminance value of the radiation CT data using the conversion coefficient.

[0016] (10) Further, the distribution calculation method of the present invention further includes a step of irradiating the target sample with radiation of a single wavelength to acquire projection image data, and a step of reconstructing the radiation CT data from the acquired projection image data.

[0017] (11) Further, the distribution calculation method of the present invention further includes a step of irradiating the target sample with radiation and acquiring projection image data by a detector having energy resolution, and a step of reconstructing the radiation CT data from the acquired projection image data.

[0018] (12) Further, the distribution calculation program of the present invention uses radiation CT data obtained by measuring a target sample formed of a material having a single composition, and performs a process of separating the region of the target sample and the region around the target sample in the radiation CT data, a process of calculating a conversion coefficient for converting the luminance value of the target sample at each pixel in the region of the target sample into a mass density, and a process of calculating the distribution of physical property values in the target sample from the luminance value of the radiation CT data using the conversion coefficient, and causes a computer to execute the processes.

Brief Description of Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Mode for Carrying Out the Invention

[0020] Next, embodiments of the present invention will be described with reference to the drawings. For ease of understanding of the description, the same reference numerals are given to the same components in each drawing, and duplicate descriptions are omitted.

[0021] [Principle] The radiation CT apparatus irradiates the sample with conical or parallel beam radiation from all angles, and the detector acquires the distribution of the radiation absorption coefficient, that is, the projection image. In order to irradiate the radiation from all angles, the radiation CT apparatus is configured to rotate the sample stage with respect to the fixed radiation source and detector, or to rotate the radiation source and detector with respect to the sample. The rotation is relative.

[0022] In this way, the distribution of the linear absorption coefficient of the sample can be inferred from the shades of the projection images of the sample obtained by performing projections from various angles. And obtaining the three-dimensional linear absorption coefficient distribution from the two-dimensional projection image is called reconstruction. Basically, backprojection of the projection image is performed.

[0023] In a radiation CT, there is the following relationship of formula (1) between the radiation absorption amount and the linear absorption coefficient μ of a substance.

Equation

[0024] Also, when monochromatic radiation passes through a uniform substance, there is the following relationship of formula (2) between the incident radiation intensity I0, the transmitted radiation intensity I, and the optical path length d.

Equation

[0025] And, there is the following relationship of formula (3) between the linear absorption coefficient μ, the mass absorption coefficient μm, and the mass density ρ of the substance.

Equation

[0026] Calculating the linear absorption coefficient μ from the radiation intensity I, and further the mass density ρ may seemingly appear easy. The mass absorption coefficient μm can be theoretically calculated from the interaction with X-rays. The mass absorption coefficient μm is a function of the wavelength and composition. However, its calculation is complex, and the mass absorption coefficient μm cannot be obtained unless the ratio of the constituent elements is clear. The linear absorption coefficient μ of a substance can be calculated by the product of its mass density ρ and the mass absorption coefficient μm.

[0027] However, the radiation intensity I at each point of the projection image data actually acquired by CT measurement is the count value of photons incident on each X-ray detection element at each position of the detector. Then, a radiation luminance value is calculated as the voxel value of the image reconstructed by CT using the projection image data. Therefore, the radiation intensity I cannot be simply replaced with the radiation luminance value to calculate the linear absorption coefficient μ. When trying to obtain the radiation luminance value from the linear absorption coefficient μ, the calculation is difficult because the detector characteristics and the radiation transmittance of the sample vary with the wavelength. The radiation luminance value is affected by, for example, the radiation spectrum, air absorption, air scattering, and the sensitivity characteristics of the detector.

[0028] The inventor noticed that even if the mass absorption coefficient and the detection efficiency for each wavelength are unknown, they can be collectively represented by a conversion coefficient α. The luminance value of the image reconstructed by CT shows the distribution of the linear absorption coefficient μ. The relationship is expressed by the following mathematical formulas (4) and (5) in terms of the luminance value CT(x, y, z), the conversion coefficient α, and the mass density distribution ρ(x, y, z).

[0029]

Equation

Equation

[0030] By using the conversion coefficient α, it is possible to calculate the distribution of the mass density from the radiation luminance value without knowing the sample composition or the mass absorption coefficient μm. Here, considering the shape of the sample on the CT image, the mass density of the structure excluding the internal voids is calculated. Also, it is assumed that the sample is composed of a single composition. If the substances have the same composition, even for a sample with an unknown composition and internal density variations, their mass absorption coefficients μm are the same. According to the present invention, it becomes possible to evaluate a gradient material with a continuously varying mass density and integrally combined with a single composition. For example, it can be applied to formed products such as resins, powders, and polymer gels. FIG. 1 shows a conceptual diagram of the cross-section of a sample with different resins (resin A, resin B) of unknown composition but a single composition and voids.

[0031] Basically, the same CT data measured from the same sample is used for calculating the physical property value distribution and the average mass density. The sample used for calculating the physical property value distribution is referred to as the "target sample". However, for calculating the average mass density, a separate sample may be cut out from the original sample of the target sample and used. The average mass density is calculated from the volume obtained from the CT data and the weight value obtained by weighing. By increasing the size of the sample piece, the weighing accuracy of the weight value is improved, so that the average weight density can be calculated more accurately. This sample used for calculating the average mass density is referred to as the "reference sample". Also, in the case of a sample for which it is difficult to calculate the average mass density from the CT data and the weight value, a literature value may be used as the reference value of the average mass density.

[0032] Also, based on the calculated mass density distribution, it is also possible to convert it into a distribution of physical property values related to the mass density of the sample, such as the filling ratio distribution, the porosity distribution, or the crystallinity distribution. Figures 2(a) and (b) are diagrams showing the procedures for calculating the filling ratio (porosity) and the crystallinity distribution of the porous body, respectively.

[0033] As shown in Fig. 2(a), by dividing the mass density distribution ρ(x, y, z) by the bulk mass density ρbulk, the filling ratio distribution f(x, y, z) can be calculated. Also, in the case of a sample with a single composition, the surroundings of the sample other than the part where the target sample is filled can be regarded as, for example, air. By subtracting the filling ratio with the entire region of the obtained CT data as 100%, the porosity distribution Pore(x, y, z) can also be calculated. By converting based on the mass density distribution, the filling ratio and porosity of structures with sizes that cannot be resolved by CT can be evaluated.

[0034] In general, it is known that crystallinity is correlated with mass density. As shown in Fig. 2(b), the distribution of mass density ρ(x, y, z) can also be converted into the distribution of crystallinity C(x, y, z). For example, for the location where the mass density is the highest (mass density ρa) and the location where the mass density is the lowest (mass density ρb), the crystallinity (crystallinity a, crystallinity b) can be obtained by an analysis method such as XRD respectively and used for the conversion. Not limited to crystallinity, as long as it is a physical property value confirmed to be correlated with density, the distribution of mass density can be converted into the distribution of the physical property value.

[0035] In the present invention, for a sample composed of a single composition with internal coarseness and fineness, the mass density is calculated by the conversion coefficient α without using a calibration sample, and the distribution of the physical property value related to the mass density is calculated from the distribution of the mass density. The conversion coefficient α is a coefficient for converting the luminance value of the target sample in each pixel into the mass density. Since the luminance value, the mass density, and the converted physical property value are added with position information (x, y, z), the number of pixels (voxel number) can be calculated for the class of the obtained or calculated value. The relationship between these values can be represented as a frequency distribution table or a histogram.

[0036] When the reconstructed data obtained by reconstructing the projection image data has a sample protruding from the CT imaging region (the region where the projection image data is acquired), the luminance value of the CT image does not match the absolute value of the linear absorption coefficient. With such an influence, the luminance value cannot be converted, so it becomes difficult to obtain the distribution of quantitative physical property values. Therefore, in order to obtain the distribution of the physical property value using the above formulas (4) and (5), it is preferable that the target sample is within the region where the projection image data is acquired. Specifically, it is preferable that the width of the sample in the direction perpendicular to the rotation axis is smaller than the width of the imaging region at any irradiation angle. Note that the width is the length in the direction perpendicular to the rotation axis at an arbitrary position on the rotation axis. When the width of the sample in the direction perpendicular to the rotation axis is larger than the width of the imaging region, it is preferable to obtain the reconstructed data by an offset scan that expands the field of view. Thereby, since the CT imaging region expands, the width of the sample does not protrude from the CT imaging region.

[0037] Also, it is preferable that a reconstruction calculation method for supplementing the condition can be selected for projection image data obtained without satisfying the condition that the target sample fits within the region where the projection image data is acquired. In addition to normal reconstruction calculations, it is preferable that a reconstruction calculation for projection images obtained by offset scanning for expanding the field of view or a reconstruction calculation considering the influence of sample protrusion (interior CT reconstruction calculation) can be selected. For example, by applying the image reconstruction method of interior CT, it becomes possible to acquire projection image data of a large object that does not fit within the field of view. In this way, by ensuring the quantitativeness of the luminance values of the CT images, it is possible to acquire the distribution of quantitative physical property values.

[0038] Also, in order to accurately calculate the distribution of physical property values, it is preferable to irradiate the target sample with X-rays of a single wavelength to acquire projection image data. For that purpose, characteristic X-rays may be used, or a multilayer mirror or crystal may be used to irradiate with X-rays of a single wavelength. By using the obtained projection image data, the sensitivity at the time of radiation detection can be improved, and the quantitativeness of the linear absorption coefficient in the radiation CT data can be improved.

[0039] The target sample may be irradiated with radiation including a wide range of wavelengths, and projection image data may be acquired by a detector having energy resolution. Radiation CT data is reconstructed from the acquired projection image data. By using the projection image data detected by a detector having energy resolution, the sensitivity at the time of radiation detection can be improved, and the quantitativeness of the linear absorption coefficient in the radiation CT data can be improved.

[0040] First, the present invention performs radiation CT measurement, separates the sample and the atmosphere on the data from the obtained reconstructed data, and calculates the average luminance values of the sample and the surroundings of the sample. In this specification, the reconstructed data is referred to as "radiation CT data" or "CT data". From the average luminance values of the sample and the surroundings of the sample obtained from the CT data and the average mass density of the sample, the conversion coefficient α is calculated by the above formula (5). Using the conversion coefficient α thus obtained, the luminance value of the CT image can be converted into the spatial distribution ρ(x, y, z) of the mass density. Hereinafter, the configuration and method for implementing the present invention will be described.

[0041] [Configuration of X-ray CT Measurement System] FIG. 3 is a schematic diagram showing the configuration of the X-ray CT measurement system 100. As shown in FIG. 3, the X-ray CT measurement system 100 includes an X-ray CT apparatus 200, a computer 300 (control apparatus and distribution calculation apparatus), an input apparatus 410, and an output apparatus 420. These apparatuses and parts are connected by wire or wirelessly so that control information, measurement data, etc. can be transmitted and received. In the present embodiment, although the use of X-rays is described as a representative example, radiation includes ionizing radiation other than X-rays, neutrons, and quantum rays.

[0042] [Configuration of X-ray CT Apparatus] The X-ray CT apparatus 200 includes a control unit 210, a stage drive mechanism 230, a sample stage 250, an X-ray generation unit 260, and a detector 270, and performs X-ray CT measurement of the held sample. The X-ray generation unit 260 has an X-ray source 265 inside. The X-ray CT apparatus 200 is for X-ray CT, and the obtained projection image data is transmitted to the computer 300.

[0043] The control unit 210 receives an instruction from the computer 300 and controls the proximity and separation of the X-ray generation unit 260 and the detector 270 with respect to the rotation center. The control unit 210 controls the rotation of the sample stage 250 according to an instruction from the computer 300. Further, the control unit 210 receives an instruction from the computer 300 and controls the acquisition of projection image data by the X-ray generation unit 260 and the detector 270.

[0044] The stage drive mechanism 230 can adjust the position of the rotation center of the sample stage 250 with respect to the X-ray source 265 and the detector 270. Further, the stage drive mechanism 230 rotates the sample stage 250 around the rotation center. The stage drive mechanism 230 can rotate the sample stage 250 at a speed set during CT measurement. Also, after the measurement is completed, the sample stage 250 can be rotated in reverse to the original position.

[0045] The sample stage 250 can place and fix the sample. The sample stage 250 is provided so as to be rotatable with respect to the X-ray source 265 and the detector 270 in order to obtain projection image data by rotational measurement. In the above description, the X-ray CT apparatus 200 is described as a stage drive type apparatus, but it may be an arm type apparatus that rotates the X-ray source 265 and the detector 270 together with the rotating arm. That is, since the measurement is performed based on the relative positional relationship between the sample, the X-ray source 265, and the detector 270, a method of moving or rotating the X-ray source 265 and the detector 270 with respect to the sample may be adopted. Also, the sample stage 250 may be configured to be attached with an attachment for controlling temperature and humidity.

[0046] The X-ray generation unit 260 and the detector 270 are basically fixed, but the distance between the two may be adjustable. The sample stage 250 is provided rotatably on the optical axis connecting the X-ray source 265 and the detector 270, with an axis provided perpendicular to the optical axis as the rotation center. Also, the sample stage 250 is provided so as to be movable together with the rotation center.

[0047] The X-ray generation unit 260 generates X-rays by the X-ray source 265 and irradiates them toward the detector 270. The X-ray generation unit 260 preferably can irradiate characteristic X-rays with wavelengths such as Cr, Cu, and Mo. The detector 270 is formed in a panel shape and has a light-receiving surface for receiving X-rays. The detector 270 detects the X-rays irradiated from the X-ray generation unit 260 and transmitted through the sample. When the X-ray generation unit 260 irradiates continuous X-rays, the detector 270 is preferably a detector having energy resolution such as a semiconductor detector. Further, the detector 270 may be provided so as to be movable in a direction perpendicular to the rotation axis. The X-ray CT apparatus 200 performs CT measurement at the calculated timing of the start of CT measurement and acquires projection image data of the sample.

[0048] [Configuration of Computer] FIG. 4 is a block diagram showing the configuration of the X-ray CT measurement system 100. In FIG. 4, mainly the functional configuration of a computer (control device and distributed calculation device) is shown. The computer 300 is, for example, a PC and is composed of a processor that executes processing, a memory or a hard disk that stores programs and data, and the like. The computer 300 includes a measurement control unit 310, a measurement data storage unit 315, an input control unit 320, an output control unit 325, a reconstruction unit 330, a CT data storage unit 335, a classification unit 340, a physical quantity data storage unit 350, an average mass density calculation unit 360, a conversion coefficient calculation unit 370, and a distributed calculation unit 380. Each unit can transmit and receive information via the control bus L.

[0049] The computer 300 functions as a control device 390 and also functions as a distributed calculation device 395. Each function is achieved by causing the computer 300 to execute a program. As shown in FIG. 4, the computer 300 receives user input from an input device 410 such as a keyboard and a mouse. On the other hand, the computer 300 displays a sample external shape image, a reconstructed image, an input screen, etc. on an output device 420 such as a display.

[0050] (Control Device) The computer 300, as a control device 390, transmits measurement conditions and the like to the X-ray CT device 200 and controls the operation of the X-ray CT device 200. The control device 390 is composed of a measurement control unit 310, a measurement data storage unit 315, an input control unit 320, and an output control unit 325.

[0051] The measurement control unit 310 adjusts the relative positions of the X-ray source 265 and the detector 270 with respect to the rotation center C0. The rotation center C0 is set between the X-ray source 265 and the detector 270, and the sample S0 is disposed at the position of the rotation center C0. The measurement control unit 310 rotates the sample around the rotation center C0 at the start of measurement.

[0052] The measurement control unit 310 acquires projection image data obtained by controlling the operations of the X-ray generation unit 260 and the detector 270. At this time, measurement is performed while rotating the sample S0. As a result, projection image data of the sample is acquired for different rotation angles over a rotation angle of 180° or more, and the measurement data storage unit 315 stores the acquired projection image data of the sample. Also, when the sample width is larger than the CT imaging region, the measurement operation is controlled to perform an offset scan. Projection images are acquired while shifting the detector or the sample rotation axis and rotating the sample.

[0053] The input control unit 320 performs control for accepting, for example, the input of CT measurement conditions from the input device 410. Also, the output control unit 325 controls the output to the output device 420 such as the display of an input screen and the display of projection image data.

[0054] (Distribution calculation device) The computer 300, as a distribution calculation device 395, processes the acquired projection image data. The distribution calculation device 395 may be separate from the control device 390 or may be placed on the cloud. The CT data is displayed by an output device 420 such as a display. The CT data storage unit 335 stores the reconstructed CT data.

[0055] The reconstruction unit 330 reconstructs three-dimensional CT data from projection image data. The CT data used for distribution calculation is preferably taken such that the image of the target sample fits within the CT imaging region in a cross-section perpendicular to the rotation axis. Such CT data can convert the luminance value into a mass density distribution using a conversion coefficient. When the width of the sample is 1 to 2 times larger than the field of view, projection images are acquired by offset scanning. It is preferable to apply an offset reconstruction operation to the acquired projection images. As a result, since the CT imaging region expands, the width of the sample does not protrude from the CT imaging region. Also, when the lateral width of the sample is larger than twice the field of view width, normal CT measurement is performed to acquire CT projection images. It is preferable to apply a sequential reconstruction operation used for interior CT image reconstruction to the acquired projection images. Thereby, the deviation between the luminance value and the linear absorption coefficient of the CT image caused by the width of the sample protruding from the CT imaging region is compensated, so that the formula for the conversion coefficient can be used.

[0056] The separation unit 340 separates the region of the target sample in the CT data from other regions. When separating, conditions based on the histogram of luminance values and position information are used.

[0057] The physical quantity data storage unit 350 stores the input information received from the user or the measuring device by the input control unit 320, the data of the average mass density calculated by the average mass density calculation unit, etc. The input information is, for example, data for converting physical property values from the mass density. It stores the bulk mass density data of the sample of the literature value and the crystallinity data acquired by an analysis method such as XRD.

[0058] The average mass density calculation unit 360 calculates the volume of the region of the separated target sample, and calculates the average mass density of the target sample using the volume calculated from the CT data and the weight value. Since it is calculated based on the CT data, it is possible to measure the actual volume even for samples with cavities inside or irregularly shaped samples. Also, instead of the CT data and weight value of the target sample, the average mass density can be calculated from the CT data and weight value of a reference sample cut out from a common sample piece. In addition, in the case of a sample for which the average mass density cannot be obtained by such a method, a literature value can be input as a reference value. The user can arbitrarily determine which average sample density to use.

[0059] The conversion coefficient calculation unit 370 calculates a conversion coefficient α based on the luminance value of the target sample using the CT data measured for the target sample formed of a material with a single composition. The conversion coefficient calculation unit 370 preferably calculates the conversion coefficient α by dividing the difference between the average luminance value of the target sample and its surroundings by the difference between the average mass density of the target sample and its surroundings. The surroundings of the target sample can be, for example, air, and the average mass density can be regarded as 0. Thereby, the conversion coefficient α can be calculated without knowing the sample composition and the mass absorption coefficient.

[0060] The distribution calculation unit calculates the mass density distribution from the luminance value of the CT data of the target sample using the conversion coefficient α. Also, the number of voxels of the CT pixels for the calculated mass density class is calculated. The calculated value may be stored in the physical quantity data storage unit 350 as a frequency distribution table. In this way, by calculating the conversion coefficient using the luminance value of the reference sample formed of a material with a single composition, the distribution of the physical property values in the target sample can be calculated from the CT data without separately preparing a calibration sample.

[0061] The distribution calculation unit calculates the distribution of physical property values that are correlated with the mass density. It converts the mass density to a physical property value based on a value for converting from the mass density, such as the bulk mass density data of the sample, to the physical property value. By storing the conversion value correlated with the mass density in the physical quantity data storage unit, it is possible to calculate the distribution of the filling rate, porosity, crystallinity, etc. of the target sample. Further, the distribution calculation unit calculates the distribution of physical property values from the CT data before and after the environmental change, and calculates the amount of change in each pixel of each data.

[0062] [Distribution calculation method] A method for calculating the distribution of physical property values using the X-ray CT measurement system 100 configured as described above will be described. FIG. 5 is a flowchart showing the distribution calculation method. Here, a method for calculating the distribution will be described using the sample of FIG. 1 as an example. The composition of the example resin A is the same as that of the resin B, and their densities are different. The periphery of the sample is air. Although the composition of air is originally different from that of the resin, here it is assumed that the void (air) inside the sample is resin C having the same composition as resin A and resin B. As a result, (CT data luminance value of resin C) - (CT data luminance value of air) = 0. At this time, the luminance values of resins A, B, and C in the obtained CT data indicate the difference in X-ray absorption from the air. Thus, it is possible to calculate the distribution of physical property values even for a sample having voids inside.

[0063] First, X-ray CT measurement is performed, and the luminance value (x, y, z) is read from the obtained reconstructed data (CT data) (step S1). The resin is formed to have a size that fits within the CT imaging region. When measuring a powder sample, for example, it is placed in a case transparent to X-rays, such as a Kapton tube, and X-ray CT measurement is performed. Such a case is treated equivalently to air. When the sample is larger than the CT imaging field of view, the projection image data obtained by offset scanning or the projection image data where the sample protrudes from the CT imaging field of view is used. For example, the reconstruction unit may determine the reconstruction calculation method from the imaging conditions of the projection image data and perform reconstruction. Alternatively, a screen for selecting the data and the reconstruction calculation method may be displayed so that the user can select arbitrarily.

[0064] Next, the sample area and the air area are separated (step S2). When separating the sample area and the air area, the separation is performed with reference to the respective luminance values and positional conditions. Next, the average luminance values of the sample and air areas are calculated respectively (step S3).

[0065] Next, the average mass density of the target sample is obtained (step S4). The method for obtaining the average mass density will be described later. For example, when the average mass density calculation unit 360 displays options such as "target sample", "reference sample", and "literature value" and the user makes a selection, the average mass density is calculated based on the corresponding CT data and the input data.

[0066] Then, the difference in the average luminance values between the sample and the air (average luminance value of the sample - average luminance value of the air) is divided by the average mass density of the obtained sample to calculate the conversion coefficient α (step S5). Using the conversion coefficient α, the luminance value CT(x, y, z) of the CT data is converted into the spatial distribution ρ(x, y, z) of the mass density (step S6). As a result, the CT image can be two-dimensionally displayed as a mass density distribution. FIGS. 6(a) and (b) are two-dimensional displays showing the distribution of the luminance value and the distribution of the mass density in the cross-section of the xy plane of the resin, respectively. Also, a three-dimensional display or a one-dimensional display in a specified direction can be performed.

[0067] Furthermore, a conversion value for converting from the mass density distribution to a physical property value related to the mass density is obtained (step S7), and the distribution of the physical property value is calculated (step S8). For example, the distribution calculation unit displays a screen on which the name of the convertible physical property value and the conversion value required for the conversion can be input. When the user selects any physical property value and inputs the conversion value, the distribution of the physical property value can be calculated. Also, if the conversion value is stored in advance, this step can be omitted, and the distribution of the physical property value can be output only by selecting the name of the physical property value. The conversion value to be input is the bulk mass density, the imaging location, or the crystallinity corresponding to the mass density. These use values obtained from the literature or other analysis methods.

[0068] The distribution of the desired physical property values is displayed as an image (step S9), and a series of steps is terminated. For example, in addition to two-dimensional and three-dimensional distribution diagrams as shown in the figure, it may be represented by a histogram in which the horizontal axis represents the mass density and the vertical axis represents the number of pixels (voxel number) with respect to the class of the mass density as shown in FIG. 7. For the physical property value selected by the user, the distribution can be displayed by three-dimensional display using a perspective view of three axes, two-dimensional display using a plane, and one-dimensional display using a graph. Further, the distributions of mass density, filling rate, porosity, and crystallinity may also be displayed as histograms.

[0069] Note that as CT data, the change amount of the distribution of the physical property values of the target sample at each pixel may be calculated using the reconstructed data of the target sample before and after the environmental change. For example, the distribution calculation unit displays a screen on which the data to be compared and the display method of the physical property value distribution can be selected. By the user selecting the data before and after the change and the desired display format, the comparison of the distributions before and after the change and the change amount of each pixel value are displayed. Thereby, the change in the distribution of the physical property values due to environmental changes such as changes in pressure, tension, temperature, humidity, and deterioration and decomposition processes over time in the non-destructive in-situ observed CT data can be measured. Further, in the above example, the filling rate and the like are calculated using the mass density, but the conversion coefficient α may be adjusted so that the filling rate and the like can be converted only by multiplying the luminance value by the conversion coefficient α.

[0070] [Method for calculating average mass density] In the above method for calculating the distribution of physical property values, the method for calculating the average mass density, which is acquired to calculate the conversion coefficient α, will be described. FIG. 8 is a flowchart showing the method for calculating the average mass density.

[0071] First, the luminance values (x, y, z) are read from the CT data used for calculating the average mass density (step T1). Next, the sample and the atmosphere are separated, and the region of the sample is extracted (step T2). When extracting the region of the sample from the data of the target sample, the result of step S2 may be used. When calculating the average mass density using a reference sample, X-ray CT measurement of the reference sample is performed in advance and the CT data is stored.

[0072] Next, the volume is calculated from the number of pixels of the CT data corresponding to the region of the extracted sample (step T3). The weight value of the sample for which the CT data was measured is measured with an electronic balance or the like (step T4). The average mass density of the sample is calculated from the volume and weight data of the sample calculated from the CT data (step T5). In this way, even if there are differences in density inside the sample, the average mass density of the sample can be calculated.

[0073] [Example 1] The above distribution calculation method was applied to a sample of ranitidine type I crystal powder. Ranitidine type I crystals were placed in a Kapton tube and CT measurement was performed. Using an X-ray source with a Cu target, 40 kV, and 30 mA, CT measurement was performed in 30 minutes at a resolution of 1.3 μm / voxel with respect to a field of view of 1.3 mm. CT reconstruction was performed by the FDK method. ImageJ was used as the image analysis software.

[0074] Figs. 9(a) and (b) are a two-dimensional X-ray CT image showing the sample and a graph showing the frequency distribution of the luminance values, respectively. Figs. 10(a) and (b) are a two-dimensional CT image showing the mass density distribution of the sample and a graph showing the frequency distribution of the mass density, respectively. Fig. 11 is a three-dimensional CT image showing the mass density distribution of the sample. As shown in Figs. 10(a), (b), and Fig. 11, the conversion of the luminance value distribution to the mass density distribution for the ranitidine type I crystal powder was successful. Also, as shown in Fig. 10(b), there are peaks at two positions with mass densities of 0.25 g / ml and 0.33 g / ml, and it was found that there are two types of particles, dense particles and sparse particles, in the ranitidine type I crystal powder.

[0075] [Example 2] The above distribution calculation method was applied to measure the filling rate of the epoxy monolith porous body. Figs. 12(a) and (b) are a three-dimensional CT image and a cross-sectional SEM photograph showing the organic porous membrane, respectively. As shown in Figs. 12(a) and (b), a large number of pores of about several μm are formed.

[0076] A Cu wire source was used to reconstruct CT data at 0.3 μm / pixel. The mass density distribution was calculated using the conversion coefficient α, and further the filling rate was calculated. FIGS. 13(a) to 13(c) are a two-dimensional CT image showing the luminance value distribution of the organic porous membrane, a two-dimensional CT image showing the filling rate distribution, and a graph showing the filling rate in the thickness direction, respectively. As shown in FIGS. 13(a) to 13(c), it was found that the filling rate is high near the interface of the porous body.

[0077] This international application claims priority based on Japanese Patent Application No. 2021-102051 filed on June 18, 2021, and incorporates the entire contents of Japanese Patent Application No. 2021-102051 into this international application.

Explanation of Reference Numerals

[0078] 100 X-ray CT measurement system 200 X-ray CT apparatus 210 Control unit 230 Stage drive mechanism 250 Sample stage 260 X-ray generation unit 265 X-ray source 270 Detector 300 Computer 310 Measurement control unit 315 Measurement data storage unit 320 Input control unit 325 Output control unit 330 Reconstruction unit 335 CT data storage unit 340 Separation unit 350 Physical quantity data storage unit 360 Average mass density calculation unit 370 Conversion coefficient calculation unit 380 Distribution calculation unit 390 Control device 395 Distribution calculation device 410 Input device 420 Output device C0 Center of rotation L Control bus

Claims

1. Using the radiation CT data obtained by measuring a target sample formed of a single-component material, a separation unit that separates the region of the target sample and the region around the target sample in the radiation CT data, a mean mass density calculation unit that calculates the volume of the region of the separated target sample and calculates the mean mass density of the target sample using the calculated volume and the weight value of the target sample obtained by weighing, a conversion coefficient calculation unit that calculates a conversion coefficient for converting the luminance value of the target sample at each pixel in the region of the target sample into a mass density using the mean mass density of the target sample, a distribution calculation unit that calculates the distribution of physical property values in the target sample from the luminance value of the radiation CT data using the conversion coefficient, A distribution calculation device comprising the above.

2. The distribution calculation device according to claim 1, wherein the conversion coefficient calculation unit calculates the conversion coefficient by dividing the difference in the average luminance value between the target sample and the region around the target sample by the difference in the average mass density between the target sample and the region around the target sample.

3. The radiation CT data obtained by measuring the target sample is The distribution calculation device according to claim 1 or claim 2, wherein the data is reconstructed from projection image data acquired under the condition that the target sample fits within the region where the projection image data is acquired.

4. The distribution calculation device according to claim 1 or claim 2, further comprising a reconstruction unit that can select a reconstruction calculation method for compensating for the condition for projection image data acquired without satisfying the condition that the target sample fits within the region where the projection image data is acquired.

5. The distribution calculation device according to any one of claims 1 to 4, wherein the distribution calculation unit calculates the distribution of any one of the filling rate, porosity, and crystallinity of the target sample from the distribution of the mass density based on a conversion value correlated with the mass density.

6. The distribution calculation device according to any one of claims 1 to 5, wherein the distribution calculation unit calculates the change amount of the physical property value of the target sample at each pixel from the distribution of the physical property value calculated from the CT data of the target sample before and after the environmental change.

7. Using the radiation CT data obtained by measuring a target sample formed of a single-component material, a step of separating the region of the target sample and the region around the target sample in the radiation CT data, Calculating the volume of the region of the separated target sample, and calculating the average mass density of the target sample using the calculated volume and the weight value of the target sample obtained by weighing; Calculating a conversion coefficient for converting the luminance value of the target sample at each pixel in the region of the target sample into a mass density using the average mass density of the target sample; Calculating the distribution of physical property values in the target sample from the luminance values of the radiation CT data using the conversion coefficient; A distribution calculation method characterized by including the above.

8. Irradiating the target sample with radiation of a single wavelength to obtain projection image data; Reconstructing the radiation CT data from the obtained projection image data, the distribution calculation method according to claim 7, further comprising the above.

9. Irradiating the target sample with radiation and obtaining projection image data using a detector having energy resolution; Reconstructing the radiation CT data from the obtained projection image data, the distribution calculation method according to claim 7, further comprising the above.

10. Using radiation CT data obtained by measuring a target sample formed of a material of a single composition, separating the region of the target sample and the region around the target sample in the radiation CT data; Calculating the volume of the region of the separated target sample, and calculating the average mass density of the target sample using the calculated volume and the weight value of the target sample obtained by weighing; Calculating a conversion coefficient for converting the luminance value of the target sample at each pixel in the region of the target sample into a mass density using the average mass density of the target sample; A distribution calculation program characterized by causing a computer to execute processing for calculating the distribution of physical property values in the target sample from the luminance values of the radiation CT data using the conversion coefficient.

Images