阅读说明：本技术 处理x射线检测数据的数据处理装置及数据处理方法及搭载有该装置或方法的x射线检查装置 (Data processing device and data processing method for processing X-ray detection data, and X-ray inspection device equipped with the same ) 是由 山河勉 宫下清哉 大杉淳 坂本恭平 早川龙太郎 于 2020-05-27 设计创作，主要内容包括：本发明提供一种数据处理装置,其向对象物照射连续X射线,并对从通过光子计数型X射线检测装置检测到的X射线透过数据获得的数据进行处理。基于所述数据针对每个检测器像素计算与对连续X射线设定的n个能量区域各自相对应的、相当于X射线透过对象物时的射线衰减值的、与该n个相对应的n维矢量。针对基于多个检测器像素中的一个或者多个检测器像素虚拟地设定的多个探索区域中的每一个,将属于该多个探索像素的各探索像素的检测器像素的n维矢量在n维空间上相互进行矢量累加。由此,针对每个该探索区域计算代表该多个探索区域的每一个的该n维代表矢量。基于每个探索区域的代表矢量、以及在具有X射线透过对象物时的该X射线的衰减程度作为坐标信息的物质空间中虚拟地设定的期望尺寸的单位区域,取得表示对象物的物质的量、种类以及性质的信息。(The invention provides a data processing device which irradiates continuous X-rays to an object and processes data obtained from X-ray transmission data detected by a photon counting X-ray detection device. Based on the data, n-dimensional vectors corresponding to n energy regions set for the continuous X-rays and corresponding to the n attenuation values when the X-rays transmit the object are calculated for each detector pixel. For each of a plurality of search regions virtually set on the basis of one or more of a plurality of detector pixels, n-dimensional vectors of detector pixels belonging to search pixels among the plurality of search pixels are vector-added to each other in an n-dimensional space. Thus, the n-dimensional representative vector representing each of the plurality of search areas is calculated for each of the search areas. Information indicating the amount, type, and nature of a substance of an object is acquired based on a representative vector for each search area and a unit area of a desired size virtually set in a substance space having, as coordinate information, the degree of attenuation of an X-ray when the X-ray is transmitted through the object.)

1. A data processing device irradiates a subject with n continuous X-rays having different energy regions, where n is a positive integer of 3 or more, and processes data indicating the degree of attenuation of the X-rays transmitted through the subject by each of a plurality of detector pixels included in a detector for each of the n energy regions,

the data processing device is characterized by comprising:

a pixel vector calculation unit that calculates, for each of the detector pixels, an n-dimensional vector corresponding to the n number of X-rays corresponding to a radiation attenuation value of the X-rays when the X-rays transmit the object, based on the data;

a representative vector calculation unit that performs vector addition of the n-dimensional vectors of the detector pixels belonging to each of the plurality of search pixels on an n-dimensional space for each of a plurality of search regions virtually set based on one or more of the plurality of detector pixels, and calculates the n-dimensional representative vector representing each of the plurality of search regions for each of the search regions; and

and a substance information acquisition unit that acquires information indicating at least one of an amount, a type, and a property of the substance of the object, based on the representative vector for each search area and a unit area of a desired size virtually set in a substance space having, as coordinate information, a degree of attenuation of the X-ray when the X-ray transmits through the object.

2. The data processing apparatus of claim 1,

each of the plurality of search areas is constituted by an area of one of the plurality of detector pixels.

3. The data processing apparatus of claim 1,

each of the plurality of search regions is configured by virtually bundling a predetermined number of mutually adjacent plurality of detector pixels among the plurality of detector pixels.

4. The data processing apparatus of claim 1,

each of the plurality of search areas is constituted by an area of discrete pixels of the plurality of detector pixels.

5. The data processing apparatus according to any one of claims 2 to 3,

the number n is three, and the number n is three,

the pixel vector calculation means is configured to calculate a three-dimensional vector as the n-dimensional vector for each of the search regions,

the representative vector calculation means is configured to calculate a three-dimensional representative vector representing the unit region in a three-dimensional space for each of the search regions,

the substance information acquiring unit is configured to acquire information indicating at least one of a type and a property of the substance based on the three-dimensional vector for each search region and the unit region.

6. The data processing apparatus of claim 5,

the substance information acquisition unit includes:

a two-dimensional declination coordinate setting unit which sets a two-dimensional declination coordinate, wherein the two-dimensional declination coordinate is respectively distributed with two declination angles of the representative vector relative to two axes on the three-dimensional coordinate; and

and a unit region setting unit configured to virtually set, as each of the plurality of unit regions, a two-dimensional unit region corresponding to an identification resolution at the time of identifying the type of the substance on the two-dimensional off-angle coordinate.

7. The data processing apparatus of claim 6,

the display device is provided with a unit area display means for displaying the two-dimensional unit area set by the unit area setting means either during the setting or after the setting, alternately.

8. The data processing apparatus of claim 6,

the substance information acquisition unit includes:

a deflection angle calculation unit that calculates the two deflection angles of the three-dimensional representative vector for each search area;

a declination determination unit that determines to which of the unit regions the two declinations of each of the three-dimensional representative vectors are classified in the two-dimensional declination coordinates;

a weight value accumulation unit configured to accumulate, for each of the unit regions, a weight value μ, which is a vector length of a corresponding three-dimensional representative vector of the three-dimensional representative vectors of the search region belonging to the unit region, as a weight value indicating a degree of X-ray attenuation for each of the unit regions_mρt，μ_mIs a mass attenuation coefficient of the object existing on an X-ray path when the X-ray transmits the object toward each of the detector pixels, ρ is a density of the object existing on the X-ray path, and t is a length on the X-ray path in the object;

a weight value image arithmetic unit that calculates a weight value image in which the accumulated weight value of each unit area is used as a pixel value; and

an analysis unit that analyzes information indicating at least one of a type and a property of the substance based on the weight value image and an accumulated value of the weight values.

9. The data processing apparatus of claim 8,

the analyzing means is configured to determine an atomic number of one element included in the substance or an effective atomic number of a substance composed of a plurality of elements, based on a pixel value of the weight value image.

10. The data processing apparatus of claim 8 or 9,

the analyzing means is configured to identify information indicating the amount of the substance indicated by the corresponding effective atomic number, based on the accumulated value of the weight values.

11. The data processing apparatus of claim 10,

the analyzing means is configured to identify information reflecting the amount (t) of the substance represented by the effective atomic number.

12. The data processing apparatus of claim 11,

the analysis unit includes: solving for the mass attenuation coefficient (μ) of the substance for each of the unit regions_m) A unit of the product with the density (ρ); and means for solving for said quantity (t) from the product.

13. The data processing apparatus of claim 11 or 12,

the analysis means is configured to indicate information reflecting the quantity (t) or a quantity processed based on the quantity (t).

14. The data processing apparatus of any one of claims 8 to 13,

the material information acquiring means is configured to record position information of each of the search regions virtually bundled in the material space, and the weight value and the pixel value in a corresponding relationship with each other.

15. The data processing apparatus of any one of claims 6 to 14,

the unit area setting means is configured to divide the two axes of the two-dimensional declination coordinate into a plurality of units at a desired width scale, and set the plurality of two-dimensional small areas specified by the desired width scale as the plurality of unit areas on the coordinate.

16. The data processing apparatus of claim 15,

the desired widths of the two axes are predetermined equally spaced widths on each axis.

17. The data processing apparatus of claim 15,

the desired widths of the two axes are unequally spaced widths predetermined on each axis.

18. The data processing apparatus of claim 17,

the unequal intervals are set at unequal intervals so that the number of the three-dimensional representative vectors classified into each of the plurality of unit regions is equalized, corresponding to the skew angle information of the plurality of elements being mapped at unequal intervals on the two-dimensional skew angle coordinate.

19. The data processing apparatus of any one of claims 5 to 18,

an air layer subtracting means for subtracting data corresponding to the X-ray attenuation amount of the air layer amount transmitted by the X-ray from the data,

the pixel vector operation unit is configured to calculate the n-dimensional vector for each of the detector pixels based on the subtracted data.

20. The data processing apparatus according to claim 19, comprising:

a correction data acquisition unit configured to acquire correction data for performing beam hardening correction of the X-ray based on a model having an effective atomic number similar to that of a specific target substance among the substances;

a correction unit that applies the beam hardening correction to the subtracted data based on the correction data,

the pixel vector calculation unit is configured to calculate the n-dimensional vector for each of the detector pixels based on the corrected data.

21. The data processing apparatus of claim 20,

when the substance can be regarded as being substantially composed of two characteristic substances, a model of an effective atomic number similar to those of the specific two target substances is set, and the correction is performed.

22. The data processing apparatus of any one of claims 6 to 21,

the analysis means is configured to perform subtraction of the three-dimensional representative vector for each of a plurality of substances constituting the object on the three-dimensional coordinates, and to separate the plurality of substances.

23. The data processing apparatus of claim 20,

a weighting unit is provided that weights the data corrected by the correction unit by a weighting coefficient for each of the three energy regions and for each of the detector pixels.

24. The data processing apparatus of claim 23,

the weighting function sets the weighting coefficient so that an S/N ratio, which is a noise amount with respect to a signal of a representative vector determined from the three energy regions, becomes maximum.

25. The data processing apparatus of claim 8,

the analysis means includes means for generating an image having pixel values corresponding to only the density ρ of the substance and the thickness t in the X-ray transmission path direction in the substance from the weight value image.

26. The data processing apparatus of claim 8,

the analysis means includes a representative vector length image generation means for generating a representative vector length image in which the length of the representative vector is set to the pixel value of each search region from the representative vector.

27. The data processing apparatus of claim 26,

provided with an original image reconstruction unit which reconstructs pixel values of the detector pixels on the basis of the data and reconstructs an original image,

the analysis unit includes:

a region-of-interest setting unit that sets a region of interest in one of the at least two images of the original image, the representative vector length image, and the weight value image; and

and a coordinated display unit that displays information indicating the position of the region of interest set in the one image on the other image in coordination with the position of the region of interest.

28. The data processing apparatus of claim 26,

the analysis unit includes:

a region-of-interest setting unit that sets a region of interest in one of the representative vector length image and the weight value image;

a coordinated display unit that displays information indicating the position of the region of interest set in the one image on the other image in coordination with the position of the region of interest; and

and an image data storage unit that stores the representative vector length image and local image data of the weight value image that form a part of the region of interest.

29. The data processing apparatus of claim 28,

an image processing unit is provided for further processing the local image data.

30. The data processing apparatus of claim 21,

the specific two object substances are bone equivalent substances and soft tissue.

31. The data processing apparatus of any one of claims 1 to 30,

the n energy regions are n energy regions set on the spectrum of one continuous X-ray,

the data indicating the degree of attenuation of the X-ray is data indicating a count value of the number of X-ray photons per unit time incident into each of the detector pixels, and is data output from an X-ray detection apparatus of a photon counting type.

32. The data processing apparatus of any one of claims 1 to 30,

the n is an integer of 2 or more,

successive X-rays belonging to each of the integer number of energy regions are irradiated from one X-ray generating device to the object in time series, or are irradiated from the integer number of X-ray generating devices to the object separately and in time series,

the data indicating the degree of attenuation of the X-ray is data indicating an integrated value per fixed time of the energy of an X-ray photon per unit time incident to each of the detector pixels or a count value per fixed time of the X-ray photon, and is data output from an X-ray detection device of an integrating type or a photon counting type.

33. The data processing apparatus of any one of claims 1 to 30,

the n is an integer of 2 or more,

in the X-ray detection device of the X-ray integration type or the X-ray photon counting type in which the X-rays irradiated from one X-ray generation device to the object and transmitted through the object are sequentially arranged from the side close to the X-ray generation device to the side far from the X-ray generation device, data corresponding to the continuous X-rays belonging to each of the integral number of energy regions is output from the integral number of X-ray detection devices in accordance with the set detection characteristics of the integral number of energy regions, respectively.

34. The data processing apparatus of claim 22,

the analysis means is configured to analyze the data received from the data receiving means,

the separation of the two substances is performed by referring to a three-dimensional representative vector representing a composite substance of the two substances formed when the X-ray transmits the two substances constituting the object, the three-dimensional representative vector being formed by smoothly connecting an origin of the three-dimensional coordinates and a curve formed by smoothly connecting distribution points on the three-dimensional coordinates arranged when the beam hardening correction is performed on a plurality of substances having different atomic numbers including the two substances.

35. The data processing apparatus of claim 34,

the analysis means is configured to analyze the data received from the data receiving means,

when one of the two substances is to be separated from the other substance, the three-dimensional representative vector of the other substance is prepared as a reference vector known in advance, the three-dimensional representative vector is projected onto the surface from the tip of the three-dimensional representative vector representing the composite substance along the direction of the reference vector, and a vector connecting the projection position on the surface and the origin is obtained as a true three-dimensional representative vector of the one substance.

36. An X-ray system integrally carrying the data processing apparatus of any one of claims 1 to 34 or separating and functionally cooperating the data processing apparatus of any one of claims 1 to 34.

37. A data processing method of processing data by a photon-counting X-ray detection device which counts the number of photons per unit time that a photon of an X-ray having an energy belonging to each of n energy regions set on a spectrum of a continuous X-ray is incident on each of a plurality of physical detector pixels provided and outputs the data representing the count value, wherein n is a positive integer of 3 or more,

an n-dimensional vector corresponding to n number of X-ray attenuation values when the X-ray passes through the object is calculated for each of the detector pixels based on the data, the n number corresponding to each of the n energy regions,

for each of a plurality of search regions virtually set on the basis of one or more of the plurality of detector pixels, vector-adding the n-dimensional vectors of the detector pixels belonging to each of the plurality of search pixels to each other in an n-dimensional space, calculating the n-dimensional representative vector representing each of the plurality of search regions for each of the search regions,

information indicating at least one of the amount, type, and property of the substance of the object is acquired based on the representative vector for each search area and a unit area of a desired size virtually set in a substance space having, as coordinate information, the degree of attenuation of the X-ray when the X-ray transmits through the object.

Technical Field

The present invention relates to a data processing apparatus and a data processing method for irradiating an object to be inspected with X-rays and processing X-ray detection data obtained by collecting X-rays transmitted through the object to be inspected, and an X-ray system incorporating the apparatus or the method.

Background

Conventionally, in X-ray examinations such as medical diagnosis and nondestructive examination using continuous X-rays, so-called spot imaging is often performed in which X-rays are irradiated onto an object and the transmitted X-rays are detected by a flat panel detector. On the other hand, in recent years, in the field of the X-ray inspection, a new inspection method has been proposed. The inspection method belongs to the following technologies: that is, an image representing the inside of the object is reconstructed from X-ray detection data collected by scanning the object with X-ray beams, or the state of the inside of the object is evaluated on an elemental (substance) level.

Conventionally, apparatuses for continuously diagnosing/inspecting the inside of a patient's body and the inside of an object are used in many fields such as X-ray imaging apparatuses, X-ray CT scanners, inline nondestructive X-ray inspection apparatuses, and the like. However, these apparatuses collect X-rays, that is, X-ray photons (photons), by integrating and collecting them step by step for a fixed time, collect, as an integral value, an amount of X-rays attenuated based on an X-ray attenuation coefficient possessed by one or more elements in the body of the patient and the object, and convert the integral value into an image or the like.

The photons of the continuous X-rays emitted from the X-ray tube have energies of various values within a range up to the X-ray energy corresponding to the tube voltage, for each particle thereof. If the energy values of the X-ray particles are different, the attenuation states received from the elements (substances) within the object are different. In view of this, in recent years, X-ray inspection using a photon counting detector having an energy discrimination function has been attracting attention, as shown in patent document 1, for example.

The X-ray inspection measures (counts) the number of photons (photons) of X-rays transmitted through an object for each of a plurality of energy regions (energy Bin), and uses the measurement result. From the viewpoint of using a plurality of energy regions, a diagnostic method called DEXA method (or subtraction method) is also the same as described in non-patent document 1.

Under the above circumstances, an X-ray inspection method described in patent document 2 has been proposed. The X-ray inspection method described in patent document 2 is applied to a configuration using a photon counting detector having the energy discrimination function described above.

Specifically, according to the configuration of patent document 2, the count value of each pixel for each of, for example, three energy regions of the X-ray irradiated from the X-ray tube (21), transmitted through the object, and detected by the photon counting type detection unit (26) is collected. An image of the Object (OB) is calculated based on the count value, and a region of interest is set on the image. Further, pixel information that becomes a background of a substance (composed of one or more elements) existing in the region of interest is removed from the image. Then, the transmission characteristic specific to the X-ray of the substance for each pixel is calculated as the specific information based on the count value for each pixel for each energy region of the X-ray in the region of interest.

As one mode of the intrinsic information, patent document 2, for example, discloses a method of calculating an X-ray attenuation amount μ for each of three energy regions (energy BINs) based on an X-ray attenuation amount μ calculated for each of the three energy regions for each of the pixels_it (μ: the ray attenuation coefficient in the X-ray path direction, t: the length of the X-ray path, i ═ 1, 2, 3) calculates a three-dimensional ray attenuation vector (μ:)₁t，μ₂t，μ₃t). The three-dimensional ray attenuation vector (mu)₁t，μ₂t，μ₃t) is further normalized to a fixed length and a three-dimensional mass attenuation vector (μ) is calculated that is a factor of not thickness t and density₁’，μ₂’，μ₃’)。

The three-dimensional mass attenuation vector (mu)₁’，μ₂’，μ₃') the degree of X-ray attenuation based on the elements present in the path is reflected in the X-ray path projected at that pixel. Thus, the three-dimensional mass attenuation vector (μ) extending from the origin₁’，μ₂’，μ₃') indicates on the three-dimensional coordinates the direction inherent to the substance (referred to as an element or a combination of elements) present in the X-ray path. That is, if the same substance or substance structure is used, the vector direction is the same, and if the noise factor is removed, the same three-dimensional mass attenuation vector is theoretically obtained. Therefore, the three-dimensional mass attenuation vector (μ) of each pixel is analyzed on the normalized three-dimensional map₁’，μ₂’，μ₃') the distribution of the distribution points at the positions of the tips, thereby providing information indicating the kind and properties of the elements (substances).

Documents of the prior art

Patent document

Patent document 1: japanese patent laid-open No. 2013-119000

Grip force document 2: international publication No. WO2016/171186A1

Non-patent document

Non-patent document 1: anli technology No.87, mar.2012, "development of dual energy mode X-ray foreign body detector"

Disclosure of Invention

Technical problem to be solved by the invention

However, the three-dimensional mass attenuation vector (μ) described in patent document 2 mentioned above₁’，μ₂’，μ₃') analysis of distribution status (μ: ray attenuation coefficients) on the normalized three-dimensional map, that is, to distinguish for each of the sets of distribution points that may be considered to be a mixture of the same one or more elements. The set of distribution points is further determined for its barycentric position, and a three-dimensional vector connecting the barycentric position with the origin of coordinates is taken as the three-dimensional mass attenuation vector representing said set.

< disappearance of Length information >

Further, according to the analysis method exemplified in patent document 2, a three-dimensional ray attenuation vector (μ) is obtained₁t，μ₂t，μ₃t) has information reflecting the length t, but requires that the length of the vector be normalized. Therefore, the length information is discarded and processed, and data indicating the amount of substance cannot be obtained from the analysis result. That is, in the case of the analytical method shown in patent document 2, the type of the estimated substance is already at the maximum limit.

< necessity of solving a problem … of large calculation processing amount for a unit area >

Further, in the case of the three-dimensional distribution map proposed in patent document 2 described above, the number of distribution points to be dotted here is usually very large in the analog amount, and therefore, if this is calculated in this state, the calculation load of the processor is very large.

In view of these circumstances, it would be desirable to provide an X-ray system that does not require the use of a three-dimensional mass attenuation vector (μ) for each pixel₁’，μ₂’，μ₃') has length information, and even when there are many distribution points, the amount of calculation in analyzing substance identification information from the distribution points is significantly reduced, and the information can be easily introduced into an actual clinical site, factory, or the like.

The present invention has been made in view of the above-described situation of the conventional X-ray inspection, and an object thereof is to provide an apparatus and a method for processing X-ray detection data, which can provide information on the type and properties of a substance constituting the whole or a part (a region of interest) of an inspection object and information indicating the amount of the substance (constituted by one element or a combination of a plurality of elements) based on X-rays with a smaller amount of computation and higher accuracy, and an X-ray inspection apparatus equipped with the apparatus.

Technical solution for solving technical problem

In order to achieve the object, the present invention provides a data processing apparatus for irradiating n (n is a positive integer of 3 or more) continuous X-rays having different energy regions to an object and processing data indicating a degree of attenuation of X-rays transmitted through the object for each of the n energy regions by each of a plurality of detector pixels included in a detector. The data processing device includes: a pixel vector calculation unit that calculates, for each of the detector pixels, an n-dimensional vector corresponding to the n number of X-rays corresponding to a radiation attenuation value of the X-rays when the X-rays transmit the object, based on the data; a representative vector calculation unit that performs vector addition of the n-dimensional vectors of the detector pixels belonging to the plurality of search pixels on an n-dimensional space for each of a plurality of search regions virtually set based on one or more of the plurality of detector pixels, and calculates the n-dimensional representative vector representing each of the plurality of search regions for each of the search regions; and a substance information acquisition device that acquires information indicating at least the amount and the type of the substance of the object based on the representative vector for each search area and a unit area of a desired size virtually set in a substance space having, as coordinate information, the degree of attenuation of the X-ray when the X-ray transmits through the object.

With this configuration, a representative vector is calculated for each search area. In this calculation, a representative vector representing the entire system is obtained by performing vector accumulation on n-dimensional vectors of detector pixels belonging to the search area, that is, performing component accumulation in each axis direction of n dimensions. Therefore, the direction of the n-dimensional vector obtained in correspondence with each detector pixel can be reflected, and the processing for acquiring substance information can be performed based on the representative vector while reflecting the length component of each n-dimensional vector as it is.

In addition, in the acquisition of the material information, a unit region of a desired size virtually set in a material space having, as coordinate information, the degree of attenuation of the X-ray when the X-ray transmits through the object is used. That is, the representative vector on the n-dimensional coordinate space is replaced with a substance space having the degree of attenuation of the X-ray as coordinate information, and the substance information is acquired as information for each unit region on the substance space. Therefore, by setting the size of the unit area to an appropriate arbitrary size, the substance information of the plurality of representative vectors in the n-dimensional coordinate space can be classified (that is, digitized into multiple values) in the substance space and processed.

In this way, the information on the amount (thickness) of the substance in the object originally possessed can be utilized by classifying the unit region in the substance space without losing the length information of the representative vector for each search region. Thereby, substance information reflecting at least the amount of the substance can be obtained. These have been considered difficult in the past and are therefore breakthrough features. If the unit regions in the same substance space as the other search regions are the same at different positions in the search regions, the amounts of the respective substances can be correlated. This makes it possible to grasp the amount of substance in each substance region in the entire search region.

In addition, it is possible to define a material space and analyze material information for each unit region on the material space while having length information representing a vector.

Therefore, the information indicating the amount of the substance in the object can be provided with a smaller amount of calculation and with higher accuracy.

This data calculation can be implemented not only as the above-described data processing apparatus but also as a data processing method having the same calculation function. Further, the present invention can be similarly implemented as an X-ray inspection system (a medical X-ray diagnostic apparatus, a nondestructive X-ray apparatus, or the like) in which such a data processing apparatus or data processing method is integrally mounted or in which such a data processing apparatus or data processing method is mounted as a unit cooperating by communication.

Drawings

In the drawings, there is shown in the drawings,

fig. 1 is a block diagram illustrating an outline of an X-ray inspection apparatus according to an embodiment of a data processing apparatus and a data processing method for implementing the present invention.

Fig. 2 is a diagram illustrating a relationship between a spectrum of continuous X-rays and energy BIN set in the spectrum.

Fig. 3 is a diagram illustrating a detector pixel and a three-dimensional vector of each pixel obtained by photon counting type detection of an energy discrimination type.

Fig. 4 is a schematic flowchart showing a procedure of obtaining the three-dimensional vector shown in fig. 3.

Fig. 5 is a functional block diagram showing the basic structure of the present invention.

FIG. 6 is a more detailed functional block diagram related to vector accumulation illustrating a portion of the basic structure of the present invention.

Fig. 7 is a diagram schematically illustrating vector accumulation showing a part of the basic structure of the present invention.

Fig. 8 is a more detailed functional block diagram relating to acquisition of substance information, which shows a part of the basic configuration of the present invention.

Fig. 9 is a diagram schematically illustrating a mesh area (unit area) set on a declination coordinate showing a part of the basic configuration of the present invention.

Fig. 10 is a diagram illustrating a method of setting a mesh area (unit area).

Fig. 11 is a diagram illustrating an example of acquisition of substance information.

Fig. 12 is a diagram illustrating another example of acquiring substance information.

Fig. 13 is a diagram obtained by materialically modeling a part of the dentition of the subject as a target object for explaining yet another example of the acquisition of the material information.

Fig. 14 is a diagram illustrating acquisition of a true three-dimensional representative vector of a hard tissue based on a model of a part of the dentition of fig. 13.

Detailed Description

Hereinafter, an embodiment of a data processing apparatus and a data processing method according to the present invention and an example of an X-ray system on which the apparatus and the method are mounted or implemented will be described with reference to the drawings.

< embodiment >

A data processing apparatus, a data processing method, and an X-ray system according to one embodiment will be described with reference to fig. 1 to 14.

The data processing device and the data processing method are applicable to an X-ray system that irradiates a continuous X-ray (also referred to as a polychromatic X-ray) having a continuous energy spectrum to a target object and detects the intensity of the transmitted X-ray by an X-ray detector.

In particular, the data processing device and the data processing method are preferably applicable to an X-ray system (for example, a medical X-ray diagnostic device or a nondestructive X-ray inspection device) that performs X-ray detection by counting the number of photons (photons) of an irradiated X-ray as X-ray intensity information while discriminating each of X-ray photons into a plurality of energy values set in advance.

Further, as an energy discrimination method, there are the following methods: that is, a method of discriminating on the receiving side according to a threshold value for a reception signal set in the X-ray detector, and a method of irradiating only an equivalent energy of X-rays having energies belonging to two or more specific energy regions determined in advance in the X-ray generator. For the latter, it is known to combine in advance one or more X-ray filters that block or transmit a specific wavelength, or to use a plurality of X-ray tubes that make target materials of the X-ray tubes different. A typical example of the latter is known as the DEXA method.

As described above, the data processing method and the data processing apparatus according to the present embodiment can be applied to any system that performs energy-discriminating X-ray detection. In particular, when photon counting X-ray detection is mounted, the accuracy is maximized. Specific examples of such a system include medical X-ray devices such as X-ray mammography devices and dental X-ray imaging devices, and X-ray devices for foreign matter examination. In particular, when photon counting X-ray detection is mounted, the accuracy is maximized.

When the data processing device and the data processing method according to the present embodiment are applied to such an X-ray system, they may be installed or mounted integrally with the system, or may be installed or mounted in a terminal that is remotely connected to the system via a communication line and performs network connection. Further, if the X-ray transmission data is detected by a system for performing energy-discriminating photon counting X-ray detection, the X-ray transmission data may be implemented by being mounted or mounted on a separate processing device.

A system for performing such energy-discriminating photon counting X-ray detection is known from international publication No. WO2015/111728 a1 and the like.

The basic structure of the X-ray system is shown in fig. 1. As shown in the drawing, the X-ray system (X-ray device) 10 includes an X-ray generator 21, and the X-ray generator 21 generates X-rays of a continuous spectrum, collimates the X-rays into a beam, and irradiates the beam onto the target space OS. The X-ray productThe generator 21 includes: an X-ray tube 22 that receives a supply of a high voltage and drives the X-ray tube 22; and a collimator 23, the collimator 23 being disposed on an output side of the X-ray tube 22, and collimating X-rays generated by the X-ray tube 22 into beam-like X-rays. The focal diameter of the tube focal point F of the X-ray tube 22 is, for example, such thatThe tube focus F can be regarded as an almost point-like X-ray source. The X-rays radiated from the tube focus F include a beam consisting of photons (Photon) having various energies (X-ray energies), and have a continuous energy spectrum corresponding to the tube voltage.

The X-ray system 10 further includes a detector 24, and the detector 24 detects a beam-like X-ray in which the irradiated X-ray is transmitted through the object OB located in the object space OS. The detector 24 has a detection layer 25 having a semiconductor (CdTe, CZT, or the like) directly below an entrance window thereof, which directly converts X-rays into an electric signal, and a pixel group in which pixels having a size of 200 μm × 200 μm are two-dimensionally arranged is formed on the detection layer 25, for example.

Of course, if a photon counting type detection structure is provided as will be described later, instead of the above-described direct conversion type semiconductor detector, a so-called indirect conversion type detector may be employed in which X-rays are converted into optical signals by a scintillator and the optical signals are converted into electrical signals by a semiconductor element.

The detector 24 further includes a layer-shaped data collection circuit 26, for example, an ASIC, which processes a detection signal of each pixel for each pixel, on the lower side of the detection layer 25. The data acquisition circuit 26 is configured as a Photon counting type circuit capable of counting the number of X-ray photons (photons) incident on the pixel group of the detection layer 25 for each pixel. In addition, the circuit divides the X-ray spectrum into a plurality of regions of X-ray energy (referred to as BINs) according to the setting of a threshold for discriminating the X-ray energy, and is capable of counting photons for each pixel in the respective energy BINs.

Therefore, from the layered data collection circuit 26, count data created by processing the electrical pulse signal in response to incidence of the X-ray photon for each pixel in each energy BIN is output as frame data (a set of count data of each pixel). For example, the frame rate may be in a range of 300fps to 6,600 fps. If the overlap phenomenon of photons incident to a pixel is removed, for example, when a photon is incident, an electric pulse is excited, and thus the count data of each pixel is count data reflecting the number of the electric pulse.

As described above, from the viewpoint of the detection processing method, the detector 24 is classified into a photon counting type (photon counting type) detector. That is, the detector 24 is configured to count the number of X-rays (Polychromatic X-rays) having a continuous energy spectrum as a set of photons having various energies, in accordance with the energy BIN (region) of the X-rays and for each pixel (one or more pixels may be used). As will be described later, as this energy BIN, for example, as shown in fig. 2, three energy BINs are set: bin is₁～Bin₃. The energy BIN: the number of bins may be three or more, four or five. Energy [ keV ]]The region below the lower threshold TH1 and the region above the upper threshold TH4 (corresponding to the tube voltage) are regions that cannot be measured or are not used. Therefore, the region between the thresholds TH1 to TH4 is divided into one (in this case, the thresholds are only TH1 and TH4) or a plurality of energies BIN. For example, as shown in fig. 2, the thresholds TH2 and TH3 are set, and three energies BIN are formed.

The above structure of X-ray irradiation/detection is proposed by international publication No. WO2015/111728 a1 and the like.

The object OB located in the object space OS is scanned by the beam-like X-rays. Therefore, any configuration may be used as long as one of the pair of the X-ray generator 21 and the detector 24 and the object OB is moved relative to the other. For example, in the case of X-ray foreign matter inspection of food or the like, the belt conveyor is disposed so as to pass through the target space OS. Thus, the object OB is placed on the belt conveyor, and the object OB is X-ray scanned. In addition, although a dental panoramic X-ray imaging apparatus is available as a medical system, in this case, the jaw of the patient as the imaging object OB is positioned in the target space OS between the X-ray generator 21 and the detector 24. In this state, the pair of the X-ray generator 21 and the detector 24 is rotated while facing each other, whereby the jaw is X-ray scanned. The same applies to an X-ray mammography apparatus as a medical system. In short, the object OB is scanned along with the relative movement between the pair of the X-ray generator 21 and the detector 24 and the object OB.

The digital measurement data output from the detector 24 is processed by a processing device mounted on the X-ray system 10 or a processing device disposed outside the X-ray system 10, which is superior in energy discrimination. This processing includes image reconstruction by a tomosynthesis method, creation of an absorption vector length image (two-dimensional image) based on the reconstructed image, and creation of a three-dimensional distribution map based on the reconstructed image. These treatments are proposed by International publication No. WO2016/171186A1 and the like.

The X-ray system 10 of the present embodiment includes a data processing device 30. As shown in fig. 1, the data processing device 30 is constituted by a microcomputer CP, for example. The computer CP itself may be configured as a computer having a known arithmetic function, and includes an interface (I/O)31 connected to the detector 24 via a communication line LN. A buffer memory 32, a ROM (read-only memory) 33, a RAM (random access memory) 34, a processor 35 having a CPU (central processing unit) 35A, an image memory 36, an input device 37, and a display device 38 are connected to the interface 31 through an internal bus B so as to be communicable with each other through the bus B. The processor 35 may also be referred to as an arithmetic unit, or the like. Instead of the CPU, an MPU (Micro-processing Unit) may be used. Of course, the ROM and RAM as the memory may be used in various known manners.

Various programs readable by a computer for measurement value correction, substance identification, and the like are stored in advance in the ROM 33. Therefore, the ROM33 includes a storage area (functioning as a non-transitory computer recording medium) 33A in which these programs are stored in advance. Further, the ROM33 includes a first storage area 33B and a second storage area 33C that store beam hardening correction data (may also be referred to as calibration data) for beam hardening correction of the measured values.

The processor 35 (i.e., the CPU35A) reads a necessary program from the storage area 33A of the ROM33 to its own work area to execute it. The processor 35 is a CPU for image processing. The buffer memory 32 temporarily stores frame data transmitted from the detector 24. The RAM34 is used to temporarily store data necessary for operation in the operation of the processor 35.

The image memory 36 stores various image data and information processed by the processor 35. The input unit 37 and the display 38 function as a man-machine interface with the user, and the input unit 37 receives input information from the user. The display 38 is capable of displaying images and the like under the control of the processor 35.

The data processing device 30 may be provided as a diagnostic device or an examination device integrated with the X-ray system 10 as described above. Further, as in the present embodiment, when the data processing device 30 is communicably connected to the X-ray system 10 via the communication line LN, it may be connected online all the time or may be communicable only when necessary. Further, the data processing device 30 may be provided in a stand-alone form. Of course, the data processing device 30 may be configured by a hardware circuit that performs pipeline processing or the like.

< data Collection and data acquisition >

Thus, according to the above-described X-ray system 10, as schematically shown in fig. 3, in each of the pixels P1, P2, P3, …, Pn of the detector 24, the calculation is based on the three energies BIN: bin is₁、Bin₂、Bin₃Corresponding attenuation coefficient mu of ray₁、μ₂、μ₃Attenuation amount mu of₁t、μ₂t、μ₃t. Here, t is the length (thickness) of the path in the object of the X-ray beam passing through the object. That is, by using the attenuation amount μ as the three physical quantities₁t、μ₂t、μ₃t are respectively regarded as a physical dimension, and a three-dimensional vector (μ) of the ray attenuation coefficient can be calculated in each pixel P1(P2, P3, …, Pn)₁t、μ₂t、μ₃t) and creating a three-dimensional vector (mu) of ray attenuation coefficients on the basis thereof₁t、μ₂t、μ₃t). The three-dimensional vector (μ) is schematically shown in a part of fig. 3₁t、μ₂t、μ₃t). In the present embodiment, the three-dimensional vector (μ)₁t、μ₂t、μ₃t) may also be referred to as a pixel vector.

Then, the three-dimensional vector (μ) is known₁t、μ₂t、μ₃t) is inclined in three dimensions (theta,see fig. 3) is specific to a substance (hereinafter, simply referred to as a substance) composed of one or more elements in the object existing along the X-ray path projected onto each detector pixel. That is, the deviation caused by the statistical noise is removed according to the kind of elements constituting the substance, the three-dimensional tilt (theta,) Take the same value. These are shown by International publication No. WO2016/171186A1 (three-dimensional map) and the like.

Further, a three-dimensional vector (μ)₁t、μ₂t、μ₃t) shows a physical quantity of information reflecting the attenuation amount of the X-ray incident on each detector pixel. That is, the greater the radiation attenuation coefficient μ, and the greater the thickness t of the substance, the greater the X-ray attenuation amount μ t. Again, these are shown by International publication No. WO2016/171186A1 (three-dimensional map) and the like.

Therefore, in the present embodiment, the data processing device 30 is inputted fromThe detector counts the number of photons output per pixel, i.e. the number of emitted photons Co₁、Co₂、Co₃As raw data necessary for performing the supplied unique substance identification processing (fig. 4, step S11). Further, the number of incident photons Cl is determined according to the number of incident photons Cl to each of the detector pixels₁、Cl₂、Cl₃And number of emitted photons Co₁、Co₂、Co₃(here, e represents an exponential function),

Co₁＝Cl₁×e(-μ₁t)

Co₂＝Cl₂×e(-μ₂t)

Co₃＝Cl₃×e(-μ₃t)

the calculation is compared with three energies BIN: bin is₁、Bin₂、Bin₃Respective corresponding X-ray attenuation mu₁t、μ₂t、μ₃t (fig. 4, step S12).

Here,. mu.₁、μ₂、μ₃Is the energy BIN: bin is₁～Bin₃The condition that the virtual average radiation attenuation coefficient (i.e., the radiation attenuation coefficient of the effective energy for each energy BIN) in (1) does not depend on the thickness t is assumed. The number of incident photons is Cl collected in a state where no object is placed₁、Cl₂、Cl₃The obtained data is usually collected in advance.

Therefore, the data processing device 30 is based on the X-ray attenuation amount μ described above₁t、μ₂t、μ₃t calculating and storing a three-dimensional vector (μ) for each detector pixel₁t、μ₂t、μ₃t) (fig. 4, steps S13, S14). The processing up to this point is also shown, for example, in International publication No. WO2016/171186A 1. The data processing device 30 is therefore functionally equipped to calculate said three-dimensional vector (μ)₁t、μ₂t、μ₃t), i.e., a pixel vector operation unit of the pixel vector.

< identification of unique substance in the present embodiment >

Data processing deviceThe device 30 can be applied to the three-dimensional vector (mu) as described above₁t、μ₂t、μ₃t) ready, for example, according to instructions from the user's interaction, the substance identification shown below is performed.

Here, the substance identification represents information indicating the type (e.g., soft tissue or hard tissue, or fat, calcium, or iron content, etc.) of an element (or a plurality of elements) constituting a substance (indicating information indicating the type (e.g., effective atomic number Zeff), information indicating a change in the property of the substance, and information indicating the amount of the substance, in addition to the information indicating the type, in the present embodiment, information indicating the amount of the substance is breakthrough which has not been achieved before, and in the past, a substance (a mixture composed of a combination of one or a plurality of elements) in a target object which has received X-ray irradiation, and information indicating a change in property (e.g., how much the substance has changed from the original elemental ratio, etc. for example, by a chemical action such as oxidation/reduction, etc., the substance is going to be putrefactive). However, these methods are not sufficient, and in addition to acquisition of information on the kind and properties of a substance, acquisition of information on the amount thereof is strongly desired.

In the present embodiment, in addition to the acquisition of information indicating the type and the property of the substance, information indicating the amount of the substance existing in the X-ray path direction is estimated with high accuracy. That is, information on the amount of the substance immobilized or dispersed in the object can also be obtained. This makes it possible to enrich the information acquired for breakthrough X-ray-based substance identification, which has not been available in the past.

As processing for this, the data processing device 30 performs a large number of representative vector operations (step S31, fig. 5) and substance information acquisition (step S32). Step S31 provides a representative vector operation means functionally constituting the representative vector operation means, and step S32 provides a substance information acquisition means functionally constituting the substance information acquisition means.

The representative vector operation is to perform vector addition of n-dimensional (e.g., three-dimensional) vectors belonging to the detector pixels of the plurality of search pixels on an n-dimensional space for each of a plurality of search regions (bundled detector pixels or one unbound detector pixel) virtually set based on one or more of the plurality of detector pixels, and to calculate a representative vector representing the n-dimensional of each of the plurality of search regions for each of the search regions.

Further, the substance information acquisition process "acquires information indicating at least one of the amount, type, and property of the target substance based on the representative vector for each search area and a unit area (also referred to as a grid area) of a desired size virtually set in a substance space having the degree of attenuation of the X-ray when the X-ray transmits through the substance as coordinate information" through the representative vector calculation.

The representative vector operation (step S31) and the substance information acquisition (step S32) will be described in detail in the following order.

< representative vector operation (step S31) >

Currently, as shown in fig. 7 a, four pixels P adjacent to each other as pixels (detector pixels) physically set in the detector 24 are₁₁、P₁₂、P₂₁，P₂₂Bundled on the signal and set as a search area EX_n，m(n is 1, 2, … p, m is 1, 2, … q: p, q are positive integers of 2 or more) (step S311). Of course, each pixel P may be₁₁(～P₂₂) Assume an exploration area EX 1.

As shown in fig. 3, at these four detector pixels P₁₁、P₁₂、P₂₁，P₂₂Sets a three-dimensional vector (μ) by the above-described preparation process for each of the two₁t、μ₂t、μ₃t), the data processing device 30 will therefore apply these three-dimensional vectors (μ)₁t、μ₂t、μ₃t) is read to the work area (step S312).

Next, the data processing device 30 performs four detector pixels P to be read out for each component of the three axes₁₁、P₁₂、P₂₁，P₂₂Respective three-dimensional vector (mu)₁t、μ₂t、μ₃t) are added to each other (step S313), whereby search area EX is obtained as a representative vector_n，mThe resultant vector Vs (see fig. 7). Thus, each pixel P is obtained and calculated_i，jThree-dimensional vector (μ) of (i, j ═ 1, 2)₁t、μ₂t、μ₃t) vector of directional components.

Next, the data processing device 30 calculates and stores the length VL of the composite vector Vs and the two angles of declination θ,(step S314). Thus, information that can uniquely define the length and the deflection angle of the synthetic vector Vs in a three-dimensional space is obtained.

These processes pass through the full heuristic EX across the detector pixels_n，mIs executed (steps S315, S316).

< acquisition of substance information (step S32) >

Next, the data processing device 30 performs a process of obtaining information on the target substance (step S32 (FIG. 8: steps S321 to S327)).

First, the data processing device 30 sets two-dimensional declination coordinates to which two declination angles θ, of the representative vector with respect to two axes are assigned on the three-dimensional coordinates,(step S321: functionally corresponds to a two-dimensional declination coordinate setting means). This two-dimensional declination coordinate is schematically illustrated in fig. 9.

Next, on the two-dimensional off-angle coordinates, a plurality of two-dimensional unit regions UR (a) corresponding to the identification resolution for identifying the type of the substance are virtually setθ_l: p, k are positive integers and p is 1, …, m, …, l is 1, …, n, …) as each of the plurality of unit regions described above(step S322: functionally corresponds to a unit area setting means). In addition, when setting the unit area, the data processing device 30 may display the two-dimensional declination coordinate on the way of setting on the display 38, and interactively adjust the unit area UR (r) (the unit area UR) with the userθ_l) The size of (2) may be set simultaneously, or a unit area UR (b) having a predetermined size may be displayedθ_l) (step S322A: functionally equivalent to a unit area display unit).

Based on the unit area UR: (θ_l) Fig. 10 (a) and (B) show an example of the setting method. The unit area UR: (θ_l) The setting may be performed interactively with the user on the storage space by the data processing device 30, or may be performed by setting in advance on the memory and reading it into the work area.

The unit region UR (shown in fig. 10 (a) as an exampleθ_l) On the horizontal axis adoptThe vertical axis is set in advance (in operation) in a memory on a two-dimensional declination coordinate of θ or set on the spot. The longitudinal axes being at equally spaced intervalsDivided so that the horizontal axis has the same or different pitch P_θIs divided. Thus, on the two-dimensional declination coordinate, the vertical and horizontal directions are equal intervals and are squareUnit area UR of rectangular or rectangular shape (b:)θ_l) Is set two-dimensionally. When a plurality of representative vectors Vs belong to the unit area UR: (θ_l) In each case, substances incident on the X-ray paths of the original detector pixels representing these representative vectors Vs can be regarded as substances composed of the same element (from the viewpoint of the effective atomic number Zeff). Therefore, the unit area UR: (θ_l) As search area EX for bundling detector pixels_n，mEach of which determines whether or not the substances can function as the same type of determiner. Then, the unit area UR (in) is defined by a unit area UR on a two-dimensional declination coordinateθ_l) The directivity of the vector indicating the substance inherently has can significantly reduce the memory area and the calculation load to be used, and can accelerate the calculation, compared with the case where the vector is indicated for each of the detector pixels.

The unit region UR (B) shown in fig. 10θ_l) With the horizontal axis:the value of (1) becomes large and the division pitch becomes largeProgressively larger, and, with the vertical axis: theta increases and the division pitch P_θAn example of gradually becoming larger. This is, for example, given in the table of the angular coordinatesWhen substances with atomic numbers Z of 7 to 13 are shown, the declination of these substancesThe distribution position of θ is not necessarily a straight line, and as shown by reference to (a) and (B) of fig. 10, the distribution position is considered to be a straight line when the atomic number is small, but is curved when the distance between the distribution points becomes narrower as the atomic number becomes larger. Therefore, in the case of the method of setting the unit region in fig. 10 (B), the division pitch P is considered to be narrowed as the atomic number becomes larger, in consideration of narrowing the interval of distribution points_θAnd gradually becomes narrower. Thus, even when the atomic number is reduced, the accuracy of determining whether or not the substances are the same kind (including substances that can be regarded as substantially the same kind) is not lowered. In addition, as an example, a substance having an atomic number Z of 7 to 13 is a plurality of substances present in the oral cavity.

In addition, the unit area UR: (θ_l) The present invention is not limited to the two-dimensional off-angle coordinate, and may be set by dividing the coordinate into two-dimensional off-angle coordinates by a straight line inclined with respect to each axis, or by a wavy curve. In this unit region, one or a plurality of three-dimensional representative vectors Vs representing the same substance are concentrated as much as possible in the same unit region, and weight values in a substance space to be described later are calculated with high precision, so that an appropriate divided region can be set in relation to the precision.

Next, the data processing device 30 calculates each of the search areas EX described above_n，mTwo angles of departure of the three-dimensional representative vector (composite vector) Vs: (θ) (step S323: functionally equivalent to a declination operation unit).

When the above operation is finished, two declination angles for each of the three-dimensional representative vectors Vs in the two-dimensional declination coordinates(θ) is classified into the unit area UR: (θ_l) Is determined based on the value of the deflection angle (step S324: functionally equivalent to the declination determination unit). For example, when there is a declination having the same or similar value: (A plurality of three-dimensional representative vectors Vs of θ) are determined to belong to the same unit area UR (r:)θ_n) (see fig. 9), and records the determination result.

Next, the data processing device 30 performs processing for each unit area UR (θ_l) A plurality of search areas EX belonging to the unit area_n，mThe respective three-dimensional representative vectors Vs are read into the work area, and the vector lengths of these three-dimensional representative vectors Vs are accumulated with each other as weight values (step S325: functionally equivalent to a weight value accumulation unit). The weight value is determined by using the weight value of mu_mScalar quantity denoted by ρ t, μ_mIs shown in each unit area UR: (θ_l) The three-dimensional representative vector Vs in (b) is a mass attenuation coefficient of the substance, ρ is a density of the substance, and t is a thickness (amount) of the substance on an X-ray path. The weight value is mu_m' ρ t represents each unit region UR, (θ_l) And information t reflecting the amount of the substance.

Next, each unit area UR is calculated (θ_l) The three-dimensional or two-dimensional weight value image in which the accumulated weight values are pixel values is recorded (step S326: functionally equivalent to a weighted value image operation unit).

That is, the data processing device 30 is capable of performing, as recording means, one process of acquiring material information by recording the position information of each search area virtually bound in the material space and the pixel value formed of the weight value in a corresponding relationship with each other. This makes it possible to easily generate a weighted value image and to easily cope with the display thereof.

Fig. 11 (a) schematically shows an example of a three-dimensional weight value image. For a two-dimensional declination angle: (θ) assigning a weight value "μ_m'ρ t' is a dimension in the height direction. The mass attenuation coefficient mu_mCan be defined as representing each unit area UR: (θ_l) Is represented by a mass attenuation coefficient mu_m' can be expressed as mu_m’＝{(W₁μ_1m)²+…+W₃μ_3m)²}^1/2And energy region BIN: bin is₁～Bin₃Corresponding W₁～W₃For the weighting coefficients, appropriate values are used. For reference, fig. 11 (a) uses a model or a virtual model having an atomic number Z of 7 to 13 as an object, and thus the weight values "μ" of the substances having the atomic numbers Z of 7, 8, 9, …, and 13_m'ρ t' is expressed as height information, i.e., information reflecting the amount of the substance.

Thus, for example, a model or virtual model with an atomic number Z of 7 to 13 is used, as in fig. 11In the case where a reference curve representing the mass attenuation coefficient μm' is obtained as shown in (B) of (a), the distribution position of the three-dimensional weight value image shown in (a) of fig. 11 is calibrated using the reference curve, and the weight value "μm" is obtained by adding the reference curve to the three-dimensional weight value image_m' ρ t ' is divided by the representative mass attenuation coefficient μm ' to obtain a three-dimensional substance image representing information of the substance amount ' ρ t ' without depending on the attenuation coefficient shown in (C) of fig. 11. From the three-dimensional material image of fig. 11 (C), the angle is deviated as the atomic number becomes largerAs θ becomes larger, the substance amount "ρ t" also becomes larger, but the rising form thereof is suppressed, and therefore, a three-dimensional substance image which is easy to observe is obtained.

Further, the three-dimensional material image shown in fig. 11 (C) is projected at an off angleOn the two-dimensional plane of θ, and the height information thereof is expressed, for example, in terms of brightness and color, to obtain a two-dimensional substance image shown in (D) of fig. 11.

In fig. 11, (a), (C), and (D) are coordinate spaces in which the type of substance includes information indicating the amount and the distribution specific to the substance is indicated, and therefore, these coordinates are also referred to as three-dimensional or two-dimensional substance spaces.

In the cases of (a) to (D) of fig. 11, it is assumed that X-ray data obtained by imaging a model having an atomic number Z of 7 to 13 is processed, but there is no continuous object in reality. Therefore, a distribution corresponding to the atomic number and the effective atomic number of one or a plurality of elements included in the substance included in the object is obtained.

In various inspections, if a means capable of measuring and grasping the substance composition of an object to be inspected, which is an object to be inspected, is used, it is also possible to express desired substance information corresponding to the purpose of the inspection.

Further, the number of passesData processing device 30, based on the weight value image and the weight value mu_mThe accumulated value of ρ t is analyzed for information indicating at least one of the type, property, and amount of the substance (step S327: analyzing means). In this analysis, information indicating at least the amount and the type of the substance among the amount, the type, and the property of the object is acquired. One mode of the analysis includes displaying the acquired information on the display 38, for example.

The analysis can also be provided in various ways.

< comparative mode >

This is, for example, the way: as a result of comparing the substance image in the two-dimensional substance space of the object obtained in fig. 11 (D) with the substance image showing temporal changes or the reference substance image determined in advance, the observed elements are drawn more clearly. In this case, the inspection target needs to be imaged under the same imaging conditions, that is, under the same imaging conditions of the X-ray tube voltage, the positional relationship with the X-ray tube, the magnification, and the like.

This example is schematically shown in fig. 12. In this figure, a reference image a captured in the oral cavity of a patient is prepared_OriginalAnd preparing a substance image B to be photographed in comparison therewith_Original。

For example, these images A_Original，B_OriginalSmoothed by a gaussian filter and a Blur (Blur) filter to obtain a filtered image a_Filter、B_Filter. Next, for the image A_Filter、B_FilterFor each pixel, A is performed_Filter/B_FilterAnd obtaining a comparative image A_Filter/B_Filter。

Based on the comparison image A_Filter/B_FilterThe amount of substance is independent of the density and mass attenuation coefficient, enabling comparison of the captured substance image from a database with various viewpoints with a reference image and a standard (average) substance distribution. Thereby, difference canThe change over time can be more significantly depicted. Further, by preparing reference images of the standard model of soft tissue + bone tissue and the standard model of bone tissue only in advance, a comparison difference with these reference images can be obtained. This enables removal of as much soft tissue as possible from the captured image of soft tissue + bone tissue, and enables more accurate rendering of the bone tissue to be observed.

< separation method >

Taking a partial image of the dentition obtained by intraoral X-ray imaging as an example, the use of the search region EX will be described with reference to fig. 13 and 14_n，mThe representative vector Vs for each search area of (1) is a method for separating soft tissue from hard tissue by a separation method. In this case, the data processing device 30 also functions as a substance separation means for performing the separation process.

As shown in fig. 13, the partial X-ray image of the dentition is roughly divided into B1: enamel fraction (classified as bone tissue), B2: tooth portion (bone tissue) + alveolar bone (bone tissue) + part of soft tissue, and, B3: alveolar bone (bone tissue) + part of soft tissue. Furthermore, the bone tissues of the parts B1, B2, B3 have different values of the effective atomic number Zeff. On the other hand, even if the soft tissues of the parts B2 and B3 have the same or approximately the same effective atomic number Zeff, there is almost no error in the analysis. This is because the length of the representative vector Vs of soft tissue is significantly smaller than that of the representative vector of hard tissue.

Therefore, the following method is employed to separate the hard tissue from the partial image of the dentition obtained by the intraoral X-ray imaging.

First, search region EX was searched for only soft tissues of an oral structure_n，mThe direction of the representative vector Vss of (a) is previously databased for reference. When creating the database for reference, a beam hardening phenomenon of the X-ray collected signal is corrected by a known method for each detector pixel using a model which is an artificial material close to the soft tissue, and a representative vector Vs is calculated based on the corrected signal.

That is, as an example, the data processing device 30 can acquire correction data for performing beam hardening correction of X-rays based on a model having an effective atomic number similar to that of a specific target substance (that is, the device 30 functions as correction data acquiring means), and perform beam hardening correction on the subtracted data based on the correction data (that is, the device 30 functions as correcting means). Thereby, the pixel vector calculates an n-dimensional vector for each detector pixel based on the corrected data.

Next, the part in the oral cavity is imaged, and each search area EX is calculated by the data processing device 30 as described above_n，mAnd temporarily storing the representative vector Vs.

The calculated representative vector Vs is expressed by a synthetic vector for bone tissue and soft tissue, that is, a vector in the longitudinal direction extending from the origin as vector information actually captured and calculated, in the three-dimensional coordinates of the attenuation amount shown in fig. 14. In this figure, the short representative vector Vss extending from the origin is a representative vector for only the reference soft tissue collected in advance.

Therefore, the representative vector Vss for reference is read from the database by the data processing device 30. Next, for example, as described above, the distribution points collected in advance are mapped on the three-dimensional attenuation coordinates using a model indicating the atomic number Z of 7, 8, 9, … 13. Next, a curve CV that smoothly connects the distribution points is calculated and mapped.

Then, the data processing device 30 calculates a plane PL (see a hatched portion) connecting the curve CV, the longitudinal direction of the synthetic vector Vs relating to the actual imaging, and the origin.

Next, an intersection KT with the curve CV is calculated by drawing a line along the surface PL from the tip of the curve CV and the synthetic vector Vs involved in the actual imaging and along the direction of the representative vector Vss of the soft tissue. That is, the representative vector Vss for soft tissue held in advance as a reference is vector-subtracted from the actually measured representative vector Vs corresponding to bone tissue and soft tissue in the three-dimensional attenuation space (which is also a kind of material space), and the actual representative vector Vsb only for the obtained bone tissue is calculated. At this time, the tip of the representative vector Vsb is on the curve CV, and therefore this position represents the effective atomic number Zeff of the substance representing the representative vector Vsb.

Therefore, the data processing device 30 is located over the search area EX_n，mThe above operation is repeated. Thus, all search areas EX_n，mCan collect the effective atomic number Zeff and the representative vector Vsb of the bone tissue in the range of the portion captured in the oral cavity. The information of the representative vector Vsb can be processed into a three-dimensional or two-dimensional material image in the same manner as described above. Further, by displaying the effective atomic number Zeff in various forms such as superimposing the effective atomic number Zeff on a partial image or by reading a change in the effective atomic number Zeff, a change in bone density or the like can be captured with high accuracy.

This is because, when performing the three-dimensional vector subtraction from "bone tissue + soft tissue" - "soft tissue" singly, the vector subtraction is performed not by the simple vector subtraction in space but by the surface on which the beam hardening correction is performed (that is, the surface PL indicated by the curve CV) expressed by the model having the atomic number as the reference. This makes it possible to estimate the direction and length of the representative vector Vsb with respect to the obtained bone tissue with higher accuracy. That is, considering that the degree of the beam hardening phenomenon on a plurality of substances differs depending on the atomic number, it is possible to estimate the change in the state of the bone tissue (bone density, etc.) and the information reflecting the amount thereof with higher accuracy.

The direction information of the representative vector Vsb with respect to the soft tissue for reference may be obtained by using the collected data of the actual patient when performing the above calculation.

In addition, in carrying out the data processing of the above-described embodiment, it is desirable to perform the process of subtracting the attenuation amount by the air layer from the detected X-ray transmission data as a pre-process, and perform the above-described substance identification using collected data for each energy region from which the noise amount by the air layer is excluded as much as possible. This preprocessing is provided as a functionally provided air layer reducing unit based on the data processing device 30.

Further, although not described in particular detail, it is necessary to perform beam hardening correction using an artificial substance having a similar composition structure to a specific one or more imaging objects from the collected data.

Further, it is desirable that the X-ray attenuation value μ obtained in units of detector pixels and energy regions is measured₁t～μ₃t times each W₁～W₃The weighting coefficients of (a) are processed.

As described above, according to the data processing apparatus, the data processing method, and the X-ray system according to the present embodiment, the representative vector is calculated for each search area. In this calculation, a representative vector representing the entire system is obtained by performing vector accumulation on n-dimensional vectors of detector pixels belonging to the search area, that is, performing component accumulation in each axis direction of n dimensions. Therefore, the direction of the n-dimensional vector obtained in correspondence with each detector pixel can be reflected, and the processing for acquiring substance information can be performed based on the representative vector while reflecting the length component of each n-dimensional vector as it is.

In addition, in the acquisition of the material information, a unit region of a desired size virtually set in a material space having, as coordinate information, the degree of attenuation of the X-ray when the X-ray transmits through the object is used. That is, the representative vector on the n-dimensional coordinate space is replaced with a substance space having the degree of attenuation of the X-ray as coordinate information, and the substance information is acquired as information for each unit region on the substance space. Therefore, by setting the size of the unit area to an appropriate arbitrary size, it is possible to classify (that is, digitize to multiple values) and process the material information of the plurality of representative vectors in the n-dimensional coordinate space in the material space.

In this way, the information on the amount (thickness) of the substance in the object originally possessed can be utilized by classifying the unit region in the substance space without losing the length information of the representative vector for each search region. Thereby, substance information reflecting at least the amount of the substance can be obtained. These have been considered difficult in the past and are therefore breakthrough features. If the unit regions in the substance space that are the same as those in the other search regions are the same at different positions in the search regions, the amounts of the respective substances can be correlated. This makes it possible to grasp the amount of the substance in each substance region over the entire search region.

In addition, a material space can be defined, and the material information can be analyzed and displayed while maintaining the length information of the representative vector for each unit region on the material space.

Therefore, the information indicating the amount of the substance in the object can be provided with a smaller amount of calculation and with higher accuracy.

In the above embodiment, various modifications can be made as follows.

When a substance to be inspected can be regarded as being substantially composed of two kinds of characteristic substances, a model of an effective atomic number similar to those of two specific target substances can be set and corrected.

Further, the beam hardening correction unit may be configured to weight the data after beam hardening correction by a weighting coefficient for each of three energy regions, which are a plurality of energy regions, and for each detector pixel. The weighting coefficient may be set so that the S/N ratio, which is the amount of noise with respect to the signal of the representative vector determined from the three energy regions, becomes maximum.

As another analysis and display method, a means may be provided for generating and displaying an image in which only the density ρ of the substance and the thickness t in the transmission path direction of the X-ray in the substance are used as pixel values from the weight value image.

Further, as another analysis/display method, a representative vector length image in which the length of the representative vector is set to the pixel value of each search region may be generated from the representative vector. In this case, the image processing apparatus includes a unit for reconstructing a pixel value of a detector pixel based on data from the detector and reconstructing an original image, and sets a region of interest in one of at least two images of the original image representative vector length image and the weight value image, and displays information indicating a position of the set region of interest in cooperation with the position of the region of interest in another image as another analysis/display method.

Further, as another analysis/display method, a region of interest may be set in one of the representative vector length image and the weight value image, information indicating the position of the region of interest set in the one image may be displayed in cooperation with the position of the region of interest in the other image, and local image data of the representative vector length image and the weight value image constituting a part of the region of interest may be stored.

Further, an image processing unit may also be provided which further processes the partial image data.

In the X-ray system according to the present embodiment, the two specific target substances are, for example, bone equivalent substances and soft tissues.

Further, in the X-ray system according to the present embodiment, the n number is an integer number of 2 or more, the continuous X-rays belonging to each of the integer number of energy regions are irradiated from one X-ray generation device to the object in time series, or the continuous X-rays belonging to each of the integer number of energy regions are irradiated from each of the integer number of X-ray generation devices to the object in time series, and the data indicating the degree of attenuation of the X-rays may be data indicating an integral value per fixed time of the energy of an X-ray photon per unit time incident on each of the detector pixels or a count value per fixed time of the X-ray photon, or may be data output from an integral type or photon counting type X-ray detection device.

Further, according to any one of claims 1 to 30, in the X-ray system, the n number is an integer number of 2 or more, and in the X-ray detection device of the X-ray integration type or the X-ray photon counting type in which the integer number of X-rays irradiated from one X-ray generation device to the object and transmitted through the object are arranged in order from a side close to the X-ray generation device to a side far from the X-ray generation device, data corresponding to the continuous X-rays belonging to each of the integer number of energy regions is output from the integer number of X-ray detection devices in accordance with the set detection characteristics of the integer number of energy regions, respectively.

< applications >

The invention has the following features: the contrast resolution of the image is high; the definition is excellent; information indicating a substance and the amount of the substance that do not exist in the past can be estimated at the same time; the precision is very high; both can be intuitively grasped through images; imaging can be performed even if the X-ray dose is the same as that of a diagnostic apparatus that performs diagnosis by image in each field at present; moreover, according to the principles of the present invention, the X-ray dose can be reduced to about 1/3 to 1/4, and the like. Therefore, it can be widely used in the field of medical equipment, the field of nondestructive inspection, the field of homeland security, and the like, and a part of the possibility of application is described below together with a simple operational effect.

< medical applications >

1) Mammography … optimization of X-ray dose based on patient exposure dose reduction and identification of breast content, increased sensitivity for detection of malignant tumors in high breast patients, and the like.

2) The chest imaging device … improves the detection rate of lung cancer based on patient exposure dose reduction, improvement of clarity and contrast resolution, thyroid abnormality detection based on iodine detection, osteoporosis detection, calcification detection, and the like as additional information.

3) The orthopedic diagnostic device … detects initial fracture, rheumatism, osteoporosis, and implant planning.

4) The intraoral imaging device … reduces the exposure dose of a patient, detects osteoporosis, examines periodontal diseases, detects inflammatory reactions, judges whether or not the patient is implanted, and makes a prognosis.

5) The dental panorama apparatus … reduces exposure dose to a patient, diagnoses osteoporosis, tests for periodontal disease, calcification of carotid artery, tests for maxillary sinusitis, and the like.

6) The contrast agent application test … shows that the exposure dose of the patient is reduced, the contrast agent is greatly reduced, functional diagnosis of the heart, liver, kidney, and the like, identification of the target, diagnosis assistance of the degree of cure, and the like, treatment planning in combination with a medical device, treatment planning in combination with a heavy metal nanoparticle contrast agent, cancer diagnosis, and the like.

7) The X-ray apparatus … for patrol reduces the exposure dose of a patient, reduces the size of the apparatus, greatly improves the diagnosis accuracy, and improves the degree of freedom of imaging.

8) The patient exposure dose of the medical and dental CT … is reduced, and the precision of the current spectral (Spectroscopic) CT is greatly improved.

9) The body composition analyzer … can be applied to sports medicine, for example, to reduce the exposure dose of a patient, to reduce the size of the patient, to greatly improve the accuracy of bone salt quantification, fat mass analysis, and the like.

10) The home X-ray apparatus … reduces the exposure dose of a patient, reduces the size, combines remote medical care with sending image information to a remote image interpretation doctor to make a diagnosis, and reduces the diagnosis load on medical care in developing countries and base hospitals in developed countries.

< non-destructive inspection application >

1) The food foreign matter inspection … has a small size, improved detection sensitivity, identification of the type of foreign matter, and dramatic expansion of the inspection target.

2) The bite check … is highly sensitive, small in size, and can check for foreign matter in food at the same time.

3) The inspection apparatus … that can be used on the sales side of a supermarket or the like can be an X-ray apparatus that is used outside the X-ray management area, a food foreign matter inspection apparatus or a spoilage detection apparatus that can be used in a place near a retail store (convenience store, supermarket or the like), or the like.

4) Analysis … of fat mass of fresh fish and meat, examination of survival state of fish and livestock, muscle mass grasp, examination of parasites, and the like.

5) Fracture inspection … in racehorse examination is to make the examination device compact, improve portability, examine initial fracture, and make the examination section more efficient by enlarging.

6) The pearl culture inspection … is small in size, high in speed, and capable of inspecting the length and shape of the pearl.

7) The inspection … for changes in properties of food products observes oxidation/reduction reactions and the like in fermentation/putrefaction processes and the like.

8) Drawing, antique, grave fresco, etc. … analyzer for paintings, composition analysis of mixed materials, etc.

9) The lithium battery inspection device … checks the internal structure, checks for storage abnormality, and the like.

10) The industrial product inspection device … checks for out of stock, assembly errors, defective products, thickness measurement, oil deterioration, and the like.

11) The metal article is inspected … for internal defects, cracks, nests, corrosion, etc.

12) The piping and tubing inspection … for cracks, wall thickness, corrosion (oxidation) inspection, etc.

13) Inspection … of emulsified liquid mixture degree of two liquids and management of liquid components.

14) Strict discrimination of multiple metals … strict discrimination/classification in the case where multiple metals are present in mixture, examination of the content of a specific metal, examination of the purity of a metal, and the like.

< safety application in China >

1) The baggage inspection … device is reduced in size, has improved detection accuracy, and can be used in airports, museums, halls, customs, stadiums, public venues, and the like.

2) Anesthesia/drug testing … high speed, high accuracy testing.

3) The banknote counterfeit detection … is a high-precision and high-speed detection.

4) The radioactive substance inspection … is generally used for inspection of dangerous substances, inspection of crops and fishes, inspection of soil, inspection of plants and fishes while radioactive substances are mixed in, and the like.

The data processing apparatus, the data processing method, and the X-ray system according to the present invention are not limited to the configurations according to the above-described embodiments and modifications, and may be combined with conventionally known configurations and the like to be further modified into various configurations without changing the scope of the claims.