X-ray imaging method for an object using multi-energy decomposition

Abstract

The method disclosed in this application uses 2D radiographic hardware implemented with scatter removal methods to perform spectral imaging or multi-energy imaging, quantitatively separating X-ray images of objects with three or more component materials into their component images. The multi-energy system can be expanded by implementing DRC multi-energy decomposition and K-edge subtraction imaging methods.

Description

Technical Field

[0001] The present invention relates generally to digital X-ray imaging, and more particularly to functional imaging values ​​for enhancing X-ray images for diagnosis, detection, image guidance, analysis and material identification and characterization, real-time tracking, time-series monitoring, and positioning of each component in an object composed of multiple components. Background Art

[0002] Large, two-dimensional semiconductor digital X-ray detector arrays are widely used in medical imaging and nondestructive testing. Typically, all image information is contained in a single projected image, where internal details of the object are obscured by overlapping components. Where multi-energy X-rays are used, quantitative analysis of the measurement data is often impossible due to the lack of accuracy required for most applications. Instead, the image data is typically used for visual display and analysis.

[0003] Typically, the human body (e.g., the chest) consists of several major substances: soft tissue (including lean tissue and fat tissue), blood vessels, heart (primarily lean tissue), lungs (primarily lean tissue with pores), bones, and in some cases, microcalcification deposits and other soft tissue structures. Each pixel in such a single image contains a mixture of all tissues, plus a random scattered component. In current digital chest imaging systems using 2D digital X-ray detectors, the contribution of each component is unknown.

[0004] It is well known that random scattering signals contribute to interference and distortion in X-ray imaging. Scattering blurs the image, reduces image contrast, and degrades image quality. Scattering can contribute 20% or more to chest imaging.

[0005] For example, in lung cancer screening, the identification and characterization of both microcalcifications and non-calcified nodules are important. Current X-ray imaging using 2D detectors is limited in its ability to diagnose lung cancer compared to computed tomography due to the inability to separate the underlying components in chest imaging.

[0006] In many applications such as human imaging, spectral imaging using 2D flat panel detectors is neither accurate nor clinically meaningful due to the presence of scattering compared to spectral imaging of CT imaging, which is a quantitative method.

[0007] The present invention is based in part on the apparatus and methods disclosed in U.S. Patent Nos. 5,648,997, 5,771,269, and 6,134,297 (Chao publication). Summary of the Invention

[0008] The present invention relates generally to digital X-ray imaging and, more particularly, to methods for digitally imaging a region of interest, such as in the human body or in a non-destructive testing environment. The present invention utilizes multiple energy devices and methods for separating an X-ray image of multiple components in a region of interest into component images from the same projected 2D image path, each component image representing at least a single physical substance.

[0009] The present invention also relates to the use of quantitative analysis methods to minimize the radiation required to identify and separate component images. For example, a spectral imaging system (such as a three-energy system) can separate four or more different components, each of which has a unique signature compared to the others, either in physical substance or spatial location, or both. Similarly, a four-energy system can separate five or more different components.

[0010] One aspect of the present invention is to include an A-space method or similar method that uses a broadband X-ray beam or multiple monochromatic X-ray beams spanning a broad X-ray spectrum and measures the transmitted spectrum with a photon counting detector or an energy sensitive detector with pulse height analysis.

[0011] The present invention relates to the use of quantitative analytical methods to determine and separate components inside or outside a region of interest from the background based on the components' unique atomic or molecular composition and microstructure or density or spatial characteristics (including size, shape, pattern, or a combination of two or more of the above characteristics).

[0012] The objects of the present invention will become apparent from the following drawings and detailed description of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] For a more complete understanding of the nature and purpose of the present invention, reference is made to the accompanying drawings, in which:

[0014] Figure 1 is a schematic diagram of the basic hardware system of the present invention;

[0015] Figure 2 is a schematic diagram of a first configuration of a hardware system employed in the present invention;

[0016] Figure 3 is a schematic diagram of a second configuration of the hardware system employed in the present invention;

[0017] Figure 4 is a flow chart of a method for decomposing a dual energy of an image into first-order approximate component images;

[0018] Figure 5 is a flow chart of a method for decomposing a dual energy of an image into second-order approximate component images;

[0019] Figure 6is a flow chart of a method for obtaining a second-order approximate microcalcification image;

[0020] Figure 7 is a flow chart of a method for obtaining a microcalcification image based on information provided by images of two different tissues;

[0021] Figure 8 is a flow chart of a method for obtaining a second order approximation of first and second soft tissue images by removing microcalcifications; and

[0022] Figure 9 is a basic flow chart of a method for performing three-energy decomposition of an image.

[0023] Figure 10 is a basic flow chart of a method for iterative dual energy decomposition of an image. DETAILED DESCRIPTION

[0024] The present invention provides a method for quantitatively separating an X-ray image of a subject (such as a chest X-ray image) into multiple component images (scatter image, lung image, bone image, blood vessel image, other soft tissue image, and microcalcification image). The present invention provides a method for separating components in a region of interest in NDT applications.

[0025] In some embodiments, the present invention employs the dual energy X-ray imaging system hardware configuration described in US Pat. Nos. 5,648,997, 5,771,269, and 6,134,297 (Chao publications).

[0026] equipment

[0027] Hardware configuration for separation of primary X-ray and scatter

[0028] refer to Figure 1 When X-rays 30 from X-ray source 12 strike object 2, a portion of X-rays 30 pass through object 2 directly to detector assembly 14 without changing their direction of propagation. These are primary X-rays 32 and convey true information about the attenuation characteristics of object 2. The remainder of X-rays 30 are randomly scattered due to interactions between X-rays 30 and the material of object 2. These are referred to as scatter 34 and distort the true information.

[0029] The present invention employs one or more configurations for separating primary X-rays from scattered light. Generally, these methods are used to remove scattered light from an image. Therefore, these methods are also referred to as scatter removal methods. However, both the primary X-ray image and the scattered light image can be useful in conveying true information about the object.

[0030] The configuration for scatter removal does not actually remove scatter. Detector assembly 14 is a single 2D detector 20 that receives both primary X-rays 32 and scatter 34. This method simply assumes that scatter 34 is present, but the amount is small enough to allow for qualitative correction, while not quantitatively accurate. In certain situations, imaging results can still be achieved. The acceptable amount of scatter 34 depends on the situation and must be determined based on analysis of the specific situation.

[0031] Another approach is to separate primary X-rays and scatter in the time domain. This takes advantage of the fact that primary X-rays 32 travel in a straight line from source 12 to detector assembly 14, taking the least amount of time to travel. Because scatter 34 does not travel in a straight line from source 12 to detector assembly 14, it takes longer to reach detector assembly 14. Thus, of the X-rays that arrive at any given detector element 28 continuously over a period of time, only the first X-ray is the primary X-ray. All others are scatter.

[0032] In this example configuration, source 12 can generate X-rays in extremely short pulses (e.g., with durations in the order of picoseconds), and detector assembly 14 is a 2D detector capable of capturing images extremely quickly, in the order of picoseconds. The captured image includes primary X-rays 32 and scatter 34 that arrive at the detector during a capture time window. If the capture window is sufficiently short, the amount of scatter 34 in the captured image is minimized. As the capture window becomes shorter, scatter 34 becomes a smaller component of the captured image.

[0033] Another configuration for separating primary X-rays from scatter is described in US Pat. No. 6,134,297. Figure 2 The detector assembly 14 is shown as a three-layer structure having a front 2D detector 22 closest to the source 12, a 2D beam selector 24, and a rear 2D detector 26. Primary X-rays 32 and scatter 34 reach and pass through the front detector 22. The beam selector 24 allows only the scatter 34 to pass through to a selected location 40 of the rear detector 26.

[0034] The beam selector 24 is implemented as an array of cylinders 42 composed of X-ray absorbing material and supported by a thin plastic sheet 44 with negligible X-ray absorption. The cylinders 42 are manufactured so that their axes are aligned with the direction of travel of the primary X-rays 32, meaning that the cylinders 42 are not parallel to each other but rather lie radially with respect to the X-ray source 12. As a result, the cylinders 42 block all X-rays directly from the X-ray source 12 within their cross-sectional area. Thus, each cylinder 42 creates a "shadow" location 40 on the rear X-ray detector 26, where the intensity of the primary X-rays 32 is essentially zero, while the intensity of the scatter 34 is substantially unaffected.

[0035] Because cylinder 42 has a finite size, a small portion of scatter 34 will be blocked by shadow location 40. However, as long as cylinder 42 is small, the blocked scatter 34 can be negligibly small. If cylinder 42 is too large or too close together, a significant amount of scatter 34 will be blocked by the remainder of rear detector 26.

[0036] The more cylinders 42 there are in the beam selector 24 , the higher the measurement accuracy at the rear detector 26 .

[0037] The material of cylinder 42 must ensure that it absorbs substantially all of the primary X-rays 32 without generating any secondary X-ray emissions or causing any additional scattering. To meet these requirements, chemical elements with a medium atomic number Z, for example, materials with a Z between 20 and 34, are preferred. Cylinder 42 can also have a multilayer structure, with a core of high-Z material and an outer layer of medium-Z material. The high-Z material absorbs X-rays most efficiently, and any secondary X-ray emissions from the core material are effectively absorbed by the outer layer without causing further secondary emissions.

[0038] The thickness or height of the cylinder 42 depends on the X-ray energy, wherein higher energy X-rays require a thicker cylinder. In lower energy X-ray imaging, such as in soft tissue imaging, the cylinder 42 can be a thin disk.

[0039] The detector assembly 14 described above is used to remove scatter 34 from the image in the following manner. A low-resolution scatter-only posterior image is read from a shadowed location 40 of the posterior detector 26. A low-resolution composite (combined primary X-ray 32 and scatter 34) posterior image is read from a selected location 46 of the posterior detector 26, where the selected location 46 of the posterior detector 26 receives both primary X-rays 32 and scatter 34, uniformly covers the entire image plane of the posterior detector 26, and is proximal to the shadowed location 40. The scatter-only image is extended to the selected location 46 by interpolation. Due to the physical properties of scatter 34, interpolation does not introduce significant errors. As long as there are a sufficiently large number of data points, the error introduced by interpolation is negligible compared to other error sources (such as statistical fluctuations in the number of X-ray photons).

[0040] The scatter-only interpolated posterior image is removed from the low-resolution synthetic posterior image to generate a low-resolution primary X-ray posterior image at the selected position 46. A low-resolution primary X-ray front image is calculated from the front detector position 48 and the rear detector selected position 46 aligned with the X-ray source 12. A low-resolution scattered front image is determined by removing the low-resolution primary X-ray posterior image from the low-resolution synthetic front image. A high-resolution scattered front image is calculated by interpolating the low-resolution scattered front image. The high-resolution scattered front image is removed from the high-resolution synthetic front image to generate a high-resolution primary X-ray image.

[0041] Another configuration for separating primary X-rays from scatter is described in detail in US Pat. Nos. 5,648,997 and 5,771,269. Figure 3 The detector assembly 14 is shown as a three-tiered structure having a front 2D detector 22 closest to the source 12, a 2D beam selector 24, and a rear 2D detector 26. A combination of primary X-rays 32 and scatter 34 reaches and passes through the front detector 22. The beam selector 24 allows only the primary X-rays 32 to pass through and reach a selected location 52 of the rear detector 26.

[0042] In the simplest configuration, the beam selector 24 is a sheet of X-ray absorbing material 54 with a large number of through-going holes 56. The holes 56 are made so that their axes are aligned with the direction of travel of the primary X-rays 32, which means that, because the X-rays are emitted essentially from a point source, the holes 56 are not parallel to each other but are aligned radially with the X-ray source 12.

[0043] Due to this alignment, the aperture 56 allows all X-rays traveling along the axis of the aperture 56 to pass through, while almost all X-rays traveling in directions slightly offset from the aperture axis are completely absorbed by the bulk material of the beam selector plate 54. Consequently, only the primary X-rays 32 reach the rear detector 26. Because the aperture 56 will always have a finite size, a small portion of the scatter 34 will reach the rear detector 26. However, as long as the size of the aperture 56 is small and the thickness of the beam selector 24 is large enough, this portion of the scatter 34 is negligibly small compared to other error sources.

[0044] Preferably, the holes 56 are as small as possible. If the holes 56 are too large, they will not be enough to prevent the scatter 34 from reaching the rear detector 26. Preferably, there are as many holes as possible in the beam selector 24. The more holes 56 there are, the higher the accuracy of the measurement at the rear detector 26.

[0045] The material of the beam selector plate 54 must ensure that all scatter 34 is absorbed and that no radiation other than the primary X-ray 32 passing through the aperture 56 , including scatter 34 and secondary emissions caused by the primary X-ray X32 or by scatter 34 , reaches the rear detector 26 .

[0046] The detector assembly 14 described above is used to remove scatter 34 from the image in the following manner. A low resolution primary X-ray image is read from selected positions 52 of the rear detector 26. The selected positions 52 are those positions on the rear detector 26 that are aligned with the apertures 56 in the beam selector plate 54. A high resolution synthetic front image is read from the front detector 22. A low resolution synthetic front image is read or calculated from the front detector positions 58 that are aligned with the selected rear detector positions 52. A low resolution scattered front image is determined by subtracting the low resolution primary X-ray rear image from the low resolution synthetic front image. A high resolution scattered front image is calculated by interpolating the low resolution scattered front image. The high resolution scattered front image is subtracted from the high resolution synthetic front image to generate a high resolution primary X-ray image.

[0047] Because only selected locations 52 on the rear detector 26 are used, an alternative configuration for the rear detector 26 is to place one or more detector cells at or in the bottom of each well 56 rather than using the entire 2D detector with most of it unused.

[0048] Alternatively, the beam selector 24 may employ a bundle of tubes having X-ray absorbing walls instead of the sheet of X-ray absorbing material 54 having the through-holes 56 .

[0049] Considerations for X-ray sources and X-ray measurement equipment

[0050] One embodiment of a multi-energy system is a basis function decomposition method (A-space method), which uses a broad spectrum X-ray source and measures the transmitted spectrum using a photon counting detector with pulse height analysis or other energy-sensitive methods using an energy-sensitive detector. Broad spectrum X-ray sources are conventional X-ray sources familiar to those skilled in the art.

[0051] In another embodiment, monochromatic X-rays or X-rays with discrete energy levels are used. Such sources are typically obtained by converting a broadband source using an energy filter (e.g., a diffraction grating such as a crystal) combined with a collimator. In some examples, a broadband X-ray source can be converted to a narrowband source by using a second target.

[0052] Laser Compton and synchrotron-based quasi-monochromatic and monochromatic light sources can also be used.

[0053] Newer X-ray source technologies including liquid metal target-based, cold cathode-based, and carbon nanotube-based X-rays, light such as LEDs, lasers, and ultrafast generated X-ray sources can all be used in the present invention.

[0054] X-ray measurement equipment includes photon-counting detectors, PMTs, and other photon-counting diodes and diode arrays, as well as energy-sensitive detectors or energy-sensitive spectrometers. In some cases, the detectors measure in the X-ray spectrum. In other cases, the detectors or X-ray spectrometers can measure in the visible spectrum, where an X-ray scintillator upstream of the detection element is used to convert the X-ray light into visible light. The latter example can be used in X-ray full-field imaging using optical methods with photodetectors or in X-ray microscopy.

[0055] Multi-energy decomposition method

[0056] Regarding the method of the present invention, the basic methods for removing scatter and for dual energy decomposition described in the Chao publication are used, and these methods are not aspects of the present invention. However, the method of the present invention includes improvements to these methods.

[0057] In order to separate three or more materials using the basic energy decomposition method, a new process is required. The present invention provides a systematic method to decompose an image of multiple components overlapping with each other into multiple component images based on physical substances.

[0058] A component here is defined as a region that perturbs X-rays differently at various spectral levels compared to different components or background in the region of interest. A single image with such a component can be visualized by using dual or multiple X-ray energy imaging methods.

[0059] One or more criteria can be set by the user or a digital program to generate a single component or multiple component images, and the criteria are typically used in artificial intelligence, neural networks (such as central neural networks or dynamic neural networks) to search a reference library to retrieve a subset database for decomposition. This process enables the use of deep machine learning and artificial intelligence known to those skilled in the art.

[0060] One aspect of the present invention is to extend the dual energy decomposition method described in the Chao publication to a multi-energy (n>2, where n is the number of energy levels) imaging system.

[0061] Yet another aspect of the present invention is to extend the material calibration method used to derive material information in the dual-energy decomposition method described in Chao's publication to a multi-energy (n>2, where n is the number of energy levels) imaging system.

[0062] Another aspect of the present invention is an alternative method other than calibration methods to derive material information values ​​through existing databases based on quantitative measurements derived from tomographic data (such as CT, MRI, PET, SPECT), stored radiological measurements associated with actual materials or synthetic or simulated material representations.

[0063] Yet another aspect of the present invention is to include dual energy based methods previously described in the prior art, including methods such as those disclosed by Chao in U.S. Patent No. 6,173,034 for material differentiation, using a first order approximation for multi-energy decomposition; and in some embodiments, including a second order approximation to image unique components (such as microcalcifications, implants, contrast marker components, or surgical or biopsy tools from background tissue) as an additional new method in multi-energy decomposition.

[0064] Calibration and database creation for material composition determination

[0065] In a preferred embodiment, the method of the present invention comprises two parts: a first-order approximation and a second-order approximation.

[0066] Before imaging the object of interest, there is a calibration and database building method for multi-energy material decomposition, which involves two processes:

[0067] Process 1 is the calibration of scatter removal at various energy levels, and process 2 is the establishment of a database for material determination using a first-order approximation (including thickness and composition). In some cases, processes 1 and 2 can be implemented by the same method, such as the dual-detector scatter removal method described in Chao's publication. However, in other cases, for example, for hardware configurations that do not involve more than one detector layer for the scatter removal method, the processes can be independent. In the latter case, in some examples, it may not be necessary to calibrate the scatter removal before imaging the sample. However, for X-ray measurements at each energy, the scatter removal method can be implemented in a post-X-ray measurement step of the object of interest.

[0068] In a preferred embodiment of the present invention utilizing a multi-energy decomposition method, a first-order approximation is sufficient. When there are regions of matter meeting specific criteria, such as regions of lesion tissue containing contrast markers or calcifications (which are significantly different from background tissue containing three or more physical substances), a second-order approximation can be used. In this case, no additional X-ray energy imaging method is used, and instead, a second-order approximation including calibration and post-imaging processing as disclosed in the Chao publication is used to derive additional component images.

[0069] In one embodiment, the first-order approximation extends and improves the scatter removal method and material decomposition method of the dual-energy system as described in the Chao publication for use in a multi-energy system with n>2. The first-order approximation uses the basic method for separating primary X-rays and scatter and the basic method for multi-energy (e.g., En energy imaging including dual-energy imaging, En=2, 3, 4... energy imaging) to separate an X-ray image of mixed components (e.g., bone, microcalcifications, and soft tissue such as blood vessels, fat, and fat-free tissue) into multiple images: scatter images of the multiple components at various energy levels (e.g., in triple-energy imaging), high-energy scatter images, medium-energy scatter images, and low-energy scatter images; and for each component, a primary X-ray image, for example, a bone tissue image, various composite or single soft tissue images, and, in the presence of microcalcifications, microcalcification (or calcification) images, or images of foreign bodies (such as implants).

[0070] Building a database of material measurements: various energy levels before imaging the sample

[0071] In one embodiment, before performing multi-energy X-ray measurement on the region of interest in the object of interest and before performing the first-order or second-order approximation, a database is established based on a database construction method.

[0072] The new material database construction method extends the calibration method described in the calibration system previously developed for dual-energy systems to systems with n>2 energies. The new database construction method allows n-energy decomposition to provide true component images. As a result, the multiple component images are a very good representation of the object.

[0073] The database includes measurements of one or more actual materials or materials comprising physical substances (similar to the materials in the sample to be measured) at different X-ray spectra (specified as n>2 energy levels), representing a composition, composite composition, or multiple overlapping compositions. The database includes X-ray measurements of static composition or structure (such as bone, muscle, or a characteristic composition or structure) at each landmark stage of a dynamic cycle, such as cardiovascular muscle motion, cardiac motion or dynamic processes, such as one or more components or composite components in a region of interest, such as cation accumulation caused by cell death or diseased tissue. Similarly, various states of one or more materials in a composition, such as tissue regions or time stamps of energy ablation zones for the same composition, including multi-dimensional characteristic static and dynamic physical properties, such as the presence or absence of material or density changes during physiological changes, and multi-dimensional dynamic physiological properties. The measured X-ray data includes data describing physical and chemical properties, including data describing various thicknesses, compositions, densities, multi-dimensional structures, or other properties that are subject to different perturbations measured by various X-ray spectra.

[0074] In some examples, portions of the database may be provided by stored data or real-time measurements. For example, components such as implants, surgical tools, chemical compounds, or any object with well-defined physical and chemical properties, or X-ray measurements of properties and dimensions, may be provided as part of the database. For example, X-ray measurements of a material at various thicknesses may be provided. In some examples, thickness measurements of one or more components, as well as geometric and dimensional measurements, may be derived in real time, using X-ray measurements at multiple energies and multiple dimensions, or measurements from sensing devices at different spatial locations known to those skilled in the art.

[0075] In addition to establishing a database by the user performing calibration steps involving pre-imaging measurements, such a database or portion of a database may also be derived from an existing database based on stored radiographic measurements associated with actual materials, including measurements of pores contained in the region of interest (if any), or quantitative measurements derived from representations of synthetic or simulated materials or pores from multi-dimensional radiographic data or tomographic data, such as CT, sinograms, MRI, PET, SPECT, optical imaging, acoustics, photoacoustics, spectroscopy, and other energy, electronic, and chemically triggered measurements.

[0076] In some embodiments, a database based on dual energy or less than n energy levels may be sufficient for energy imaging-based material decomposition methods with n > 2. In some examples, relevant physical property and size information about the composition or region of interest may be known or can be derived and simulated, so such a database may not be needed.

[0077] The quantitative relationship between the sensing elements of each detector used in the system, especially with the same illumination projection path Those associated with the measurements of the diameter

[0078] In one embodiment, when two or more detector layers or sensing elements are used, such as those disclosed by Chao above, or any two or more sensing elements along the same X-ray projection path, the database can include measurements from a calibration method that establishes a database of quantitative relationships between the front detector and the rear detector. In some cases, only at selected locations, such as (i, j) and (i', j'). When the front detector and the rear detector are quite similar or essentially identical, such a calibration method relates measurements about the front detector (e.g., primary X-ray measurements) to measurements about the rear detector, which can be simplified by using existing algorithms to characterize the relationship between the front detector and the rear detector. When relevant data exists for the type of detector used, this relationship information can be simulated or derived for scatter removal purposes without the need for a calibration step.

[0079] Step 1: First-order approximation

[0080] The method for obtaining a first-order approximation consists of the following eight steps:

[0081] (a) Calibration is performed on material M. In breast imaging, for example, the first component M1 is a stent or implant, M2 is bone, M3 is soft tissue, M4 is blood vessels, and M5 is calcification.

[0082] M1=M1(D1...D n )

[0083] M2=M2(D1...D n )

[0084] M n =M n (D1...D n )

[0085] For simplicity, dual energy imaging of the chest is used for illustration. As described below, in order to obtain a pair of numerical relationships for the front detector of microcalcification or calcification c and soft tissue s at high energy and low energy, a function c=c(D H , D L ) and s=s(D H , D L ), additional energy levels can be applied to other tissues. For example, the equations are c = c(D1, D2, D3), s = s(D1, D2, D3), b = b(D1, D2, D3), ... l = l(D1, D2, D3), where b represents the equation for bone and l represents the equation for the contrast-labeled tissue. D H 、D L Denote D1 and D2, respectively. However, for simplicity, only two components and two energy approaches are shown. This dual-energy approach can be extended to multi-energy systems. The following describes the basic functions of the linearization. Therefore, if the number of unknowns corresponds to the number of equations, the basic principles are similar.

[0086] (b) Perform calibration to obtain a pair of numerical relationships between the first component f and the second component g at two energies. In this example, the function f=f(D H , D L ) and g=g(D H , D L ).

[0087] It should be noted that some or all of the required calibration can be replaced by a database derived from previous measurements and other forms of simulated data such as multi-energy radiography, CT tomography, sinograms in CT, and other forms of simulated data such as MRI, single photon emission CT, PET, spectroscopy, and optical measurements as described above. In Chao's disclosure, the calibration for scatter removal, which involves the derivation of the scalar relationship between the front and rear detectors, is performed together with the calibration for material decomposition, which provides information based on thickness and composition. In the present invention, these two calibrations can be separated and the database for material decomposition lookup can be performed as a separate method from the scatter removal method (e.g., in the case of a single detector, a single energy based scatter removal method). For illustration, a dual detector method and hardware are described. However, other non-dual detector based scatter removal methods may also be used.

[0088] (c) Irradiating an object with X-rays of energy levels H and L.

[0089] (d) Obtain high-resolution image D from the previous detection position (x, y) fHh (x, y) and D fLh (x, y), where the image includes primary X-rays and scattered X-rays.

[0090] (e) Calculate a pair of high-resolution scattered X-ray images D fSHh (x, y) and D fSLh (x, y).

[0091] (f) Calculate a pair of high-resolution primary X-ray images D fPHh (x, y) = D fHh (x, y)-D fSHh (x, y) and D fPLh (x, y) = D fLh (x, y)-D fSLh (x, y).

[0092] (g) Using the function c=c(D H , D L ) and s=s(D H , D L ), for image pair D fPHh (x, y) and D fPLh (x, y) performs dual energy decomposition to obtain two first-order approximate material composition images c1(x, y) and s1(x, y).

[0093] (h) Using the function f=f(D H , D L ) and g=g(D H , D L ), for image pair D fPHh (x, y) and DfPLh (x, y) performs dual energy decomposition to obtain two first-order approximate material composition images f1(x, y) and g1(x, y).

[0094] The present invention includes utilizing the above-mentioned database or calibration method.

[0095] In one embodiment, the present invention includes a second order approximation step.

[0096] Before imaging the object, a quantitative relationship algorithm can be established. To establish such a database, a calibration method for a second-order approximation step is used. The calibration method of the present invention comprises the steps of: (1) determining the dual energy equations for two known materials u and v as, D H =D DH [u,v],D L =D DL [u, v]; (2) Use the standard least squares data fitting method to perform functional decomposition of the energy-related attenuation coefficient function μ u (E)=u p ×μ p (E)+u q ×μ q (E) and μ v (E) = v p ×μ p (E)+v q ×μ q (E), to obtain the constant u p 、u q 、v p and v q ; (3) For each coordinate pair (u, v) calculate p = u × (u p +v p ) and q=v×(u q +v q ), to obtain the dual energy equations D H =D DH [p,q],D L =D DL [p, q]; and (4) solving the material p and q as a variable pair (D H , D L ) is a function of the equation system D H =D DH [p,q],D L =D DL [p, q], to obtain the equation system p=p(D H , D L ) and q=q(D H , D L ).

[0097] Although the first-order approximation provides a reasonably good image of the components, improvements are still possible.

[0098] For example, in lung imaging, since the chest consists of three or more tissue components, simple dual-energy decomposition has limitations. Some component images obtained in the first-order approximation still contain a small amount of mixed signals and are not pure single component images.

[0099] In a second order approximation, in the case of chest imaging, considering the multi-material composition of the chest would yield images of microcalcifications or foreign elements in the lungs, soft tissue including fat and fat-free tissue, and bone.

[0100] Isolation of unique and rare components (DRC)

[0101] In this specification, a DRC is distinct from the rest of the region of interest. Examples include microcalcifications or calcified areas, implants, and contrast marker components. A unique component not only has a unique physical composition and size compared to other regions of interest, but also differs in some key characteristics. Examples of key characteristics include, in the case of contrast marker components or microcalcifications, a lack of tissue change mitigation properties. A unique component may be relatively rare and smaller in size compared to other components in the region of interest. A DRC component can be embedded in a multi-component region where one or more tissues have similar but different X-ray measurement properties to the DRC component, for example, similar to atomic Z in soft tissue. One or more such distinguishable properties can distinguish the DRC component from the rest of the components in the region. DRC components can be corrected using a second-order approximation.

[0102] The method for obtaining a second-order approximation corrects for the effects of DRC components and comprises the steps of: (a) identifying all DRC components, or in this example, microcalcification points c1(x,y) in the image c1(x,y) k ,y k ) and all non-DRC component identification points c1(x i ,y i ); (b) construct a background image B(x, y), where point B(x i ,y i )=c1(x i ,y i ), and point B(x k ,y k ) from around point c1(x k ,y k ) of point c1(x j ,y j) for interpolation; (c) remove the background image B(x, y) from the image c1(x, y) to obtain a second-order approximate unique identification image c2(x, y); (d) identify all zero points c2(x0, y0) and non-zero points c2(x0, y0) in the image c2(x, y) m ,y m ); (e) constructing a second-order approximate image f2(x, y) of a first specific tissue (such as fat or fat-free tissue or soft tissue), wherein the point f2(x0, y0) = f1(x0, y0), and the point f2(x m ,y m ) from around point f1(x m ,y m ) of the point f1(x n ,y n ) interpolation; (f) constructing a second-order approximate image g2(x, y) of a second specific tissue, such as bone tissue, wherein the point g2(x0, y0) = g1(x0, y0), and the point g2(x m ,y m ) from around point g1(x m ,y m ) of the point g1(x n ,y n ) to perform interpolation.

[0103] Extension of the first-order approximation of the dual energy decomposition

[0104] Figure 4 The steps for a first-order approximation are shown in the flow chart of The first step is to perform two calibrations for different material pairs under the same system conditions.

[0105] The first calibration is to obtain a pair of numerical relationships for the front detector at high energy level H and low energy level L for two materials c and s, where c represents the same X-ray attenuation coefficient μ as a function of X-ray energy as that of the microcalcified deposits. c (E), and s represents the X-ray attenuation coefficient μ as a function of X-ray energy that is the same as that of ordinary tissue matter. s (E) Materials.

[0106] The following is a description of how dual energy calibration is performed for real tissue components, which are now extended to exploit, and in some cases improve, multi-energy imaging in the present invention.

[0107] The Chao publication describes the dual energy calibration method in detail. It can be extended to multiple energy system calibration when multiple energy systems are present. In the present invention, calibration can be performed based on actual material compositions.

[0108] To perform calibration measurements, according to the method described in Chao's publication, several material plates containing two synthetic materials should be used. Currently, such standard tissue samples are not available. The present invention provides an improved multi-energy calibration method that can perform calibration measurements using another pair of materials to obtain accurate calibration data for the desired different material pairs.

[0109] The scientific basis for the improved calibration method is described below.

[0110] Assume that a numerical equation for a calibration data material pair (p, q) is required, but direct material (p, q) data is not available due to some practical limitations. On the other hand, another material pair (u, v) can be used for calibration measurements. Moreover, as a function of X-ray energy μ u (E), μ v (E), μ p (E) and μ q All X-ray attenuation coefficients of (E) are known. In chest imaging, the (u, v) material pair can be, for example, a pair of commercially available materials such as polyethylene (CH2) sheets and water (H2O). More materials can be used for additional components, for example, commercially available materials such as aluminum and acrylic.

[0111] Note that the material pairs in the present invention can be two or more. Furthermore, even in a multi-energy system, the number of known materials used in the material set for the second-order approximation can be between 2 and n, and the corresponding energy levels used in the second-order approximation can also be between 2 and n. For example, microcalcifications in the lungs can be separated from the sternum, lung tissue, and other soft tissues. In this case, using multi-energy imaging to separate the microcalcifications from the lung tissue, there can be only three known material sets u, v, and w that require calibration. The first relationship set is the microcalcification c, the lung plus soft tissue composite l, and the bone b. The second relationship set is the bone b, the lung l, and the other soft tissue s.

[0112] The amount of known material may vary and the amount of material may be chosen as separate compositions for calibration and second order approximation, but the basic principles are similar.

[0113] In another embodiment, when a 2D projection image contains measurements of the same component type or similar material from two locations along the projection path, for example, measurements of a bone region overlapping a microcalcification region on a projection image in chest imaging, the margin between the overlap and non-overlap regions may have a measurable difference in bone density or thickness measurements. Microcalcification quantification can be achieved by statistically estimating bone density measurements of the region surrounding the overlap. Because the non-overlap and overlapping regions vary slowly and are therefore similar in density and size, the bone density and size measurements of the overlapping region can be interpolated based on the adjacent regions. Measurements of the microcalcification region can then be derived.

[0114] Alternatively, if multi-dimensional images are acquired and synthesized, regions with bone images and regions with microcalcification images can be distinguished from each other, and accurate density and thickness measurements can be derived.

[0115] Alternatively, for example, in imaging microcalcifications in chest imaging, the amount of microcalcifications can be estimated using a dual-energy method that diverts the X-ray image of the selected region away from the irradiation path that does not pass through the bony regions of the body. In this case, the dual-energy first- and second-order approximations required to consider only the lung tissue and other soft tissue along the projected X-ray path can be used.

[0116] In another example, in breast imaging, both tumor markers and microcalcifications can be separated using two energies. Using a three-energy approach, breast tissue can be separated into regions of fat, lean tissue, and tumor markers. When calibrated using three known materials, u, v, and w, and at three different energy levels, solving for the composite of calcification, tumor, fat, and fat-free tissue, as well as the composite of calcification, tumor, and fat-free tissue, provides additional diagnostic information.

[0117] Returning to the example of the dual energy system, the dual energy equations are written as

[0118]

[0119]

[0120] where Φ 0H (E) and Φ 0L (E) are the energy spectra of the X-ray source 14 at the higher energy level H and the lower energy level L, respectively. The projected mass densities u(x, y) and v(x, y) are expressed in grams per square centimeter (g / cm 2 ) is the unit. Mass attenuation coefficient μ u (E) and μ v (E) in square / g (cm 2 All these values ​​are known. f (E) is the X-ray spectral sensitivity of the front detector.

[0121] Energy-related attenuation coefficient function μ u (E) and μ v (E) Perform function decomposition

[0122] μ u (E)=u p ×μ p (E)+u q ×μ q (E) (2a)

[0123] μ v (E) = v p ×μ p (E)+v q ×μ q (E) (2b)

[0124] Among them, the constant u p 、u q 、v p and v q can be obtained by standard least squares data fitting methods. It should be noted that this function decomposition is generally not true from a mathematical point of view. However, it has been established through extensive analysis of experimental data that for substances including chemical elements with low to medium atomic numbers Z, such as carbon, hydrogen, oxygen, nitrogen, up to calcium in the human body, and as long as the X-ray energy is within the medical diagnostic energy range, such as between 20 keV and 150 keV, the third X-ray mass attenuation function can usually be written as the sum of the other two X-ray mass attenuation functions with good approximation. The energy-dependent attenuation function in the exponent exp() of equations (1a) and (1b) is denoted as UF(E):

[0125]

[0126] in

[0127] u(x, y)×u p +v(x, y)×v p ≡p(x, y) (4a)

[0128] and

[0129] u(x, y)×u q +v(x, y)×v q ≡q(x,y) (4a)

[0130] Among them, the symbol “≡” represents the definition, and then

[0131] -UF(E)=μ p (E)×p(x, y)+μ q (E)×q(x,y) (5)

[0132] Therefore, the improved calibration process includes the following steps:

[0133] (1) As described in Chao's publication, calibration measurements were performed using (u, v) material pairs. As a result, the numerical equations were obtained:

[0134] D H =D DH [u,v] (6a)

[0135] D L =D DL [u,v] (6b)

[0136] (2) By using the standard least squares data fitting method to fit the energy-dependent attenuation coefficient function μ u (E) and μ v (E) Perform function decomposition and obtain the constant u p 、u q 、v p and v q .

[0137] μ u (E)=u p ×μ p (E)+u q ×μ q (E) (7a)

[0138] μ v (E) = v p ×μ p (E)+v q ×μ q (E) (7b)

[0139] (3) For each data pair (u, v) in formulas (6a) and (6b), calculate

[0140] u×u p +v×v p =p (8a)

[0141] u×u q +v×v q =q (8b)

[0142] And obtain a new dual-energy numerical equation system

[0143] D H =D DH [p, q] (9a)

[0144] D L =D DL [p, q] (9b)

[0145] (4) Solve the equation system D for the variable pair [p, q] by numerical inversion H =D DH [p, q] and D L =D DL [p, q] as a variable pair [D H , D L ] function and obtain

[0146] p=p[DH , D L ] (10a)

[0147] q=q[D H , D L ] (10b)

[0148] In order to perform two types of dual energy decomposition, two corresponding types of calibration are performed. The first type of calibration is for the material pair (c, s) of the DRC component (in this case, microcalcifications) and the other type of tissue in the region of interest (in this case, soft tissue). By using the calibration process described above, a pair of quantitative explicit numerical functions is established:

[0149] D H =D DH [c,s] (11a)

[0150] D L =D DL [c,s] (11b)

[0151] Among them, the symbol D H [] and D L [] represents a functional relationship using c and s as variables. The numerical equation system is obtained by solving the equation system by numerical inversion, for example

[0152] c=c[D H , D L ] (12a)

[0153] s=s[D H , D L ] (12b)

[0154] This numerical equation set is applicable to all normalized pixels. For any pixel, the measured signal pair (D H , D L ), the material composition data pair c and s of the pixel can be easily found.

[0155] The second calibration is to obtain a pair of numerical relationships of the detector of two materials f and g at high energy level H and low energy level L, where f represents the same X-ray attenuation coefficient u as the first tissue or first component in the region of interest. f (E) and g represents a material having the same X-ray attenuation coefficient u as the second tissue or the second component. g (E) materials. Therefore, a quantitative explicit function is established:

[0156] D H =D DH [f, g] (13a)

[0157] DL =D DL [f, g] (13b)

[0158] Solve the system of equations by numerical inversion, for example, to obtain the numerical system of equations

[0159] f=f[D H , D L ] (14a)

[0160] g=g[D H , D L ] (14b)

[0161] This set of numerical equations is applicable to all normalized pixels.From the measured signal pair (DH, DL), the material composition data pair (f, g) can be easily found.

[0162] To avoid misunderstandings, two points need to be mentioned. First, for higher accuracy, one should choose a calibration material that has a composition as close as possible to the actual tissue composition in terms of X-ray interaction. This should be clear based on the general error theory of experimental measurements. For example, when μ u (E) Close to μ p (E) and μ v (E) Close to μ q (E), the error due to the transformation of (7a) and (7b) is smaller.

[0163] The second point is that there are many subtle permutations for the order of the computational steps. For example, after step (1), by performing calibration measurements using the (u, v) material pair and establishing the numerical relationships (6a) and (6b), rather than performing the material pair transformations (8a) and (8b) to obtain the numerical relationships (9a) and (9b), one can directly invert (6a) and (6b) to obtain:

[0164] u=u[D H , D L ] (15a)

[0165] v=v[D H , D L ] (15b)

[0166] Then, relations (15a) and (15b) are applied to the image data D fPHh (x, y) and D fPLh (x, y), we can obtain two physical images u(x, y) and v(x, y). Then, by using equations (8a) and (8b)

[0167] p(x, y) = u(x, y) × u p +v(x, y)×vp (16a)

[0168] q(x, y) = u(x, y) × u q +v(x, y)×v q (16b)

[0169] The entire image u(x, y) and v(x, y) is converted point by point into two images p(x, y) and q(x, y). This is merely a slight change in the order of the mathematical calculations, without any fundamental change in the method. Therefore, it can be expected that the process described in equations (1a), (1b) through (12a), (12b) includes these and other similar slight substitutions.

[0170] The second step of the method of the present invention is to obtain a pair of high spatial resolution composite images D of the front detector. fHh (x, y) and D fLh (x, y), where subscript f represents the image of the front detector, subscript H represents the high-level image, subscript L represents the low-level image, subscript h represents the high-resolution image, and (x, y) represents the two-dimensional Cartesian coordinates of the detector unit of the front detector. Image data pair D fHh (x, y) and D fLh Each image in (x, y) contains a scattered component and a main X-ray component. The main X-ray component contains a first component, a second component, and a DRC component.

[0171] The third step is to transform the image into fHh (x, y) and D fLh Each image of (x, y) is separated into a scattered component and a main X-ray component. As described in Chao's publication, this is accomplished through a large number of data acquisition and computational steps to compute a pair of high spatial resolution scattered-only images D at the front detector. fSHh (x, y) and D fSLh (x, y), where the subscript S represents the scattered image only. This step is the same as the corresponding step in Chao's publication.

[0172] The fourth step is to calculate the high spatial resolution primary X-ray image pair from the equation

[0173] D fPHh (x, y) = D fHh (x, y)-D fSHh (x, y) (17a)

[0174] D fPLh (x, y) = D fLh (x, y)-D fSLh (x, y) (17b)

[0175] The subscript P represents only the primary X-ray image. This step is also the same as that described in Chao's publication.

[0176] The fifth step is to perform image alignment D by using the calibrated numerical equations (12a), (12b). fPHh (x, y) and D fPLh Dual energy decomposition of (x, y). As a result, two material composition images c1(x, y) and s1(x, y) are obtained. In this image pair, c1(x, y) is essentially the DRC component image, and s1(x, y) is essentially the composite material image of the first and second components.

[0177] The sixth and final step is to perform image alignment using the calibrated numerical equations (14a), (14b). fPHh (x, y) and D fPLh (x, y). The result is two material composition images, f1(x, y) and g1(x, y). In this image pair, f1(x, y) is essentially the first component image, and g1(x, y) is essentially the second component image. However, both f1(x, y) and g1(x, y) may contain DRC component information that has not yet been clearly specified and properly quantified.

[0178] At this point, usable results have been obtained by decomposing a single image into the following five component images: Scattering image D fS (x, y), DRC component image c1(x, y), image s1(x, y) of the combined material with the first and second components, first component image f1(x, y), and second component image g1(x, y). However, the capabilities of dual-energy decomposition have not yet been fully utilized. The above results can be considered a good first approximation for general imaging data. The reason these component images can only be considered first-order approximations is that the dual-energy decomposition process assumes that the region of interest contains only two materials. In reality, the region of interest contains three materials.

[0179] Extension of the second-order approximation of dual-energy decomposition to multi-energy systems: DRC imaging

[0180] The next part of this specification describes a second order approximation method that takes into account three or more material components, for example in chest or heart imaging or in imaging of objects comprising three or more components. First, the scientific principle is explained and then the steps of the method are described.

[0181] In the embodiments described in the specification where a dual energy method may be used, a three-material image decomposition method based on images obtained with two different energies of X-rays may have the advantages of lower radiation levels and faster acquisition times.

[0182] In the case where there are more components t>n+1 (where n is the available energy level), the current method can also be used to separate additional components from other components.

[0183] The following explains how the n-energy method can be used to approximately separate n+1 or more components within a region of interest (ROI) in medical or industrial imaging. This is possible because specific conditions in medical imaging can be exploited to achieve a good approximation. The first specific condition is that the X-ray attenuation of at least two of the n components is very similar when compared to the n+1th component, the DRC component. The second condition is that DRC components, such as microcalcified deposits or calcified areas, implant areas, or contrast-marked lesions, are typically present in relatively small, sometimes isolated, regions and lack the slowly varying characteristics of the tissue or components found in the rest of the region of interest. For example, microcalcifications will be confined to a few individual image pixels.

[0184] For example, when n = 2, the dual energy imaging equation of the region of interest can be expressed by the equation pair

[0185]

[0186]

[0187] The projected mass densities c(x,y), f(x,y), and g(x,y) of the object are expressed in grams per square centimeter (g / cm 2 ) is the unit. Mass attenuation coefficient μ c (E), μ f (E) and μ g (E) is known and is expressed in square centimeters per gram (cm 2 / g) is expressed in units. The energy-related exponent of the exponential function exp() in the equations (18a) and (18b) is denoted as AF(E), and is

[0188]

[0189] There are three independent mass decay functions μ c (E), μ f (E) and μ g (E), which is connected by three unknown parameters c(x, y), f(x, y) and g(x, y) for the material density value. Dual energy measurement only provides two measurement signals D fPHh (x, y) and D fPLh (x, y), from which only two unknown parameters can be determined.

[0190] The basic method used to quantitatively assess each tissue component is the function decomposition method. An energy-dependent mass attenuation function must be expressed as the sum of two other functions. For example, to perform a basic dual energy decomposition of the first and second components, the DRC component function (e.g. calcium function) μ c (E) decomposes into

[0191] μ c (E)=R cf ×μ f (E)+R cg ×μ g (E) (20)

[0192] Among them, R cf and R cg is a constant. Note that for the human body system, when the X-ray energy is in the medical diagnostic energy range, the third X-ray mass attenuation function can always be written as the sum of the other two X-ray mass attenuation functions with good approximation. R can be calculated by using the standard least squares data fitting method cf and R cg For example, in the X-ray energy range of 16keV to 100keV, by applying the least squares data fitting method to equation (20), it is possible to find that R cf ≈-34 and R cg ≈33. R cf and R cg The exact value of depends on the X-ray energy range used in the hardware system.

[0193] You can use the parameter R cf and R cg Some general observations are made. First, R cf Always negative, R cg Always positive. Second, R cf and R cg The absolute values ​​of are always large and always close to each other. For example, for the diagnostic energy range, it can be assumed that R cf and R cgThe absolute value of is between 10 and 50. The underlying physical reason for these two features is that calcium has a greater attenuation coefficient at lower energies; and at low energies, various types of soft tissue generally have similar attenuation coefficients, slightly greater than that of fat. These features produce a very unusual result in the decomposed X-ray component images: when the image is decomposed into separate soft tissue component images, each DRC component region (such as an implant or microcalcification) produces a positive, high-intensity attenuation change in the first component image (such as the fat-free tissue image), and each DRC component region produces a negative, high-intensity attenuation change in the second component tissue image at that point. These positive and negative points have a precise quantitative relationship. For example, one microgram of microcalcification produces a positive attenuation change corresponding to an intensity change of approximately 30 micrograms of fat-free tissue, and a negative attenuation change corresponding to an intensity change of approximately 30 micrograms of the first soft tissue. In the present invention, these features are used to quantitatively separate the superimposed microcalcification image from the first soft tissue component image and from the second soft tissue component image.

[0194] AF(E) can be rewritten as a function consisting of only two basic functions:

[0195]

[0196] or

[0197] -AF(E)=μ f (E)×fc(x,y)+μ f (E)×gc(x,y) (22)

[0198] in

[0199] fc(x,y)=f(x,y)+R cf ×c(x,y) (23a)

[0200] gc(x,y)=g(x,y)+R cg ×c(x,y) (23b)

[0201] Since AF(E) consists of only two independent functions μ f (E) and μ g (E), so by using the dual energy decomposition process described in Chao's publication, two unique energy-independent parameters can be obtained by solving the equations (18a) and (18b), and the solution pair is fc(x, y) and gc(x, y). Note that the solution pair fc(x, y) and gc(x, y) is exactly the solution pair of the first approximation. That is,

[0202] f1(x,y)=fc(x,y)=f(x,y)+R cf ×c(x,y) (24a)

[0203] g1(x,y)=gc(x,y)=g(x,y)+R cg ×c(x,y) (24b)

[0204] After performing dual energy decomposition according to the first-order approximation process, the third material component is projected and superimposed on the two basic component images, as shown in (24a) and (24b), with R on the fat tissue image. cf ×c(x, y) and R on the fat-free tissue image cg ×c(x, y). Generally, since only fc(x, y) and gc(x, y) can be calculated by using dual energy decomposition, the portion of the superimposed third component image cannot be separated from the two component images.

[0205] The present invention utilizes specific conditions of certain features of medical imaging to approximately separate a third material composition. For example, in lung imaging, microcalcification deposits are known to exist only in small areas. For small areas, it can be assumed that the soft tissue density varies approximately smoothly. The following is an overview of the basic steps for separating microcalcification image spots from soft tissue images. First, identify the microcalcification spots on the image. Microcalcification spots must be: (a) positive signal / negative signal pairs, each pair having approximately equal absolute amplitude values ​​but opposite signs; and (b) superimposed at exactly the same corresponding positions on the first soft tissue component image and the second soft tissue component image. Second, the image intensity is interpolated from the nearby surrounding soft tissue pixels to replace the image intensity of these suspected microcalcification spots. Third, the interpolated signal intensity is removed from the original signal intensities gc(x, y) and fc(x, y) to determine whether the intensity change at the identified spots is consistent with the intensity change predicted by the decomposition equation (20) of the microcalcification function. If the microcalcification intensity on both soft tissue images is quantitatively consistent with the intensity predicted by equation (20), then the speckle satisfies the preliminary test for microcalcification speckle. Similar methods can be used for other DRC component measurements.

[0206] In addition to this preliminary testing, there were further tests as described below.

[0207] The following describes the scientific basis for obtaining pure DRC component images through the second decomposition.

[0208] Taking microcalcification measurement as an example, in the second decomposition of the first-order approximate accuracy level, two basic functions are selected as μ c (E) and μ sa (E), where μ c (E) is the energy-dependent attenuation function for microcalcifications, and μ sa(E) is the attenuation function for the average soft tissue composition. Two images c1(x, y) and s1(x, y) are obtained as a result of the second decomposition in the first order approximation. In the first order approximation, it is assumed that there are only two types of mass attenuation coefficient functions μ c (E) and μ sa (E) Material composition. According to the actual soft tissue material composition, the energy-related index in equations (18a) and (18b) is expressed as BF(E):

[0209]

[0210] or

[0211]

[0212] Among them, the average tissue mass attenuation coefficient function μ sa (E) = 50% μ f (E) +50%μ g (E). δμ s (E)×δs(x,y) is the difference between the average soft tissue attenuation μ at pixel (x, y) and the average soft tissue attenuation μ at pixel (x, y). sa (E)×s a The local deviation of (x, y). For example, when the tissue composition at (x, y) is exactly 50% fat tissue and 50% fat-free tissue, then δμ s (E) = 0. Otherwise, according to the basic function μ c (E) and μ sa (E) To perform dual energy decomposition, the energy-related function δμ must be s (E) is decomposed into two basic functions μ c (E) and μ sa (E),

[0213] δμ s (E)=R sc ×μ c (E)+R ss ×μ sa (E) (27)

[0214] Among them, R sc and R ss are two constants whose exact values ​​can be calculated using standard least squares data fitting methods when the local tissue composition is known. Then,

[0215]

[0216] or

[0217] -BF(E)=μ c (E)×cs(x,y)+μsa (E)×ss(x,y) (29)

[0218] in

[0219] cs(x,y)=c(x,y)+R sc ×δs(x,y) (30a)

[0220] ss(x, y) = s a (x, y) + R ss ×δs(x,y) (30b)

[0221] Some parameters can be used to calculate R sc and R ss First, R sc is much smaller than 1.0, and R ss The underlying physical reason is that the function δμ is s Energy-related behavior of (E) and the average organization function μ sa (E) is not much different, but is similar to μ c (E) is very different.

[0222] Note that the dual energy decomposition result c1(x, y) obtained at the first-order approximate accuracy level is exactly cs(x, y) and s1(x, y) is exactly ss(x, y).

[0223] c1(x,y)=cs(x,y)=c(x,y)+R sc ×δs(x,y) (31a)

[0224] s1(x,y)=ss(x,y)=s a (x,y)+R ss ×δs(x,y) (31b)

[0225] Therefore, due to the local deviation of the actual tissue composition from the assumed average tissue composition, the microcalcification image c1(x, y) obtained by the first-order approximation consists of two images: one is the true microcalcification image c(x, y), and the other is the image generated by the local deviation of the tissue composition from the assumed average composition.

[0226] The exact signal intensity of the superimposed image is R sc ×δ s (x, y). Generally speaking, such superimposed images due to the presence of the third material cannot be separated by the dual energy method. However, based on the use of some specific conditions of the region of interest, the present invention provides a method for removing the third material R sc ×δ sAn approximate method for the influence of (x, y) on the true third material image c(x, y).

[0227] In order to separate the third material influence on the microcalcification image, the image R sc ×δ s Two feature differences between (x, y) and image c(x, y): (a) soft tissue composition and average composition R sc ×δ s The local deviation of (x, y) has a smooth spatial distribution in a similar manner to the soft tissue spatial distribution sa(x, y), while the microcalcification image has a sharp intensity change in a small area; and (b) the deviation image R sc ×δ s The magnitude of (x, y) is small because R sc Very small.

[0228] Figure 6 The following describes a process for obtaining an image of pure third material (eg, microcalcifications) from the second decomposition at a second approximate level of accuracy.

[0229] (1) In the microcalcification image c1(x, y), first identify all microcalcification points (x k ,y k Microcalcifications are distinct from other image points because the microcalcification image c1(x, y) essentially consists of isolated, high-intensity image points against a smoothly varying background. Microcalcifications are identified by the sharp change in pixel intensity at the isolated point. Sometimes, microcalcifications can take the form of clusters, so these points are not always individually isolated. The present invention contemplates the existence of clusters containing a small number of microcalcifications, where the signal intensity of a microcalcification point is much higher than that of neighboring points.

[0230] (2) By pixel (x i ,y i )≠(x k ,y k ) constructs the background image B(x, y) for c1(x, y). That is, for all non-microcalcification points, x i ≠x k And y i ≠y k , background image B(x i ,y i )=c1(x i ,y i ). B(x i ,y i ) have excluded microcalcifications, so they must be caused by local tissue composition deviations and are therefore microcalcification background signals.

[0231] (3) In the background image B(x, y), for those pixels (x k ,y k ), assuming that the signal intensity is equal to the average signal intensity value interpolated from the signal intensities of surrounding pixels. That is, for j around k,

[0232] B(x k ,y k )=averaged[c1(x j ,y j )] (32)

[0233] Since tissue composition varies relatively smoothly compared to microcalcified deposits, the average c1(x j ,y j )(averaged c1(x j ,y j )) will be very close to the point (x k ,y k ) at the actual tissue composition changes.

[0234] (4) Remove the tissue background image B(x, y) from the first-order approximate microcalcification image c1(x, y) to obtain the second-order approximate microcalcification image

[0235] c2(x,y)=c1(x,y)-B(x,y) (33)

[0236] In image c2(x, y), the image signal intensity should accurately represent the quantitative microcalcification density without being affected by changes in tissue composition.

[0237] Figure 7 The construction of microcalcification image c based on first-order approximate fat and fat-free tissue images f1(x, y) and g1(x, y) is described in fg The steps for (x, y) are as follows:

[0238] (1) Identify points (x, y) in the first soft tissue image f1 (x, y) and the second soft tissue image g1 (x, y) where the intensity changes sharply relative to the adjacent points. m ,y m ). Point (x m ,y m ) is the potential calcification site given by the tissue images f1(x, y) and g1(x, y).

[0239] (2) Verify that each point (x m ,y mOnly when the predicted signal changes on the first soft tissue image f1(x, y) and the second soft tissue image g1(x, y) can be quantitatively determined in the new third image, in this case, the microcalcification image signal c is constructed. fg (x, y). To do this, first calculate the signal strength f of a pair of interpolated 1a (x m ,y m ) and g 1a (x m ,y m ), so that for n around m, f 1a (x m ,y m )=averaged[f1(x n ,y n )]; and for n around m, g 1a (x m ,y m )=averaged[g1(x n ,y n )]. Then calculate

[0240] Δf=-[f1(x m ,y m )-f 1a (x m ,y m )] (34)

[0241] as well as

[0242] Δg=g 1a (x m ,y m )-g1(x m ,y m ) (35)

[0243] When these two quantities are close to a certain range, for example, between 20 and 40, the pixel (x m ,y m ) calcification at the point. When Δf and Δg are inconsistent in quantity, the most likely reasons are: (a) the point selected according to the criterion of sharp intensity change is not a calcified deposit, but actually represents some unique soft tissue structure; or (b) the point may contain calcification, but the signal intensity of the calcification is too low and immersed in noise. In both cases, the selected point cannot be regarded as a calcified deposit and should be removed from the calcification image based on the information provided by the first soft tissue image f1(x, y) and the second soft tissue image g1(x, y). Therefore, the calcification image c fg(x, y) may not be exactly the same as the calcification image c2(x, y). Even though there may be slight differences between the two calcification images obtained from different data analysis methods, c2(x, y) and c fg (x, y) both provide their own reference data. Both images are available.

[0244] (3) According to the result of step (2), let

[0245] c fg (x m ,y m )=-[f2(x m ,y m )-f1(x m ,y m )]×R cf (36)

[0246] or

[0247] c fg (x m ,y m )=[g2(xm,ym)-g1(x m ,y m )]×R cg (37)

[0248] Because these values ​​are close to each other, either value from equation (36) or (37) can be used as the calcification image intensity. Choosing one as the final result is largely a matter of operator preference and depends on the operator's judgment. If further comparison is needed, the value from equation (36) may be better than the value from equation (37) because it is assumed that the signal intensity of the first soft tissue has a smoother distribution than the signal intensity of the second soft tissue. Therefore, the interpolated average value can be more accurate.

[0249] exist Figure 8 The steps of obtaining a second-order approximation of the first soft tissue image f2(x, y) and a second-order approximation of the second soft tissue image g2(x, y) from which microcalcifications are removed are described as follows:

[0250] (1) For the second-order microcalcification image c fg For each pixel at (x, y), identify those points c where the intensity is zero fg (x0, y0) and those points c whose intensity is non-zero fg (x m ,y m ).

[0251] (2) Construct images f2(x, y) and g2(x, y), where there is no change between the first-order and second-order images at the zero-intensity point, that is, f2(x0, y0) = f1(x0, y0) and g2(x0, y0) = g1(x0, y0).

[0252] (3) The remaining points f2(x m ,y m ) and g2(x m ,y m ) is set to the average of the signal strengths of nearby points, i.e., for n around m, f2(x m ,y m )=averaged[f1(x n ,y n )], and for n around m, g2(x m ,y m )=averaged[g1(x n ,y n )]. After traversing all pixels (x, y), two corrected tissue images f2(x, y) and g2(x, y) in which the influence of microcalcification is corrected are obtained.

[0253] In addition, the above method is not limited to imaging of lung or heart or vascular tissue, but can be used to image microcalcifications or calcifications in other organs, such as but not limited to the brain, gastrointestinal system, abdomen and liver. In particular, intracranial tumors contain various levels of calcifications.

[0254] Calcification patterns in pulmonary nodules (PNs) may aid in the diagnosis of lung lesions. Vascular calcification is an active and complex process involving multiple mechanisms that cause calcium deposition in arterial walls. These mechanisms lead to increased arterial stiffness and pulse wave velocity, which in turn contribute to increased cardiovascular morbidity and mortality.

[0255] Alternatively, instead of microcalcifications or calcification materials, the imaging method can be applied to DRC materials, for example, a third tissue or modified tissue, a third component labeled with a contrast agent. The contrast agent can be a nanoparticle or particle of an atom Z that is different from the other atoms in the region of interest, or a particle derivative that is bound to a molecular marker of the cell or tissue or object of interest to be imaged, isolated or quantified. The contrast agent can also be an iodinated agent, barium sulfate, or a derivative of such molecules with other existing CT or X-ray markers. The third material can also be an implant, surgical or biopsy tool that has X-ray absorption or X-ray detection properties that are different from the background.

[0256] First-order and second-order approximations for dual-energy decomposition extended to multi-energy systems

[0257] Multi-energy decomposition method

[0258] In order to separate more materials or components in an object than previously possible, imaging methods using three or more energy levels are used in a multi-energy system. The multi-energy decomposition method of the present invention uses a linearization method or a dual energy decomposition method, which is essentially an extension of those methods described in Chao's publication. Generalized spectral imaging is described in "Image Quality Metrics for Digital Systems" in "Handbook of Medical Imaging" edited by JT Doubbins III, 2000, pp. 161-219.

[0259]

[0260] where Φ 0n (E) is the energy spectrum of the X-ray source at the nth energy level En, μ n (E) is the mass attenuation coefficient of the nth material component and is expressed in square centimeters per gram (cm 2 / g) is expressed as a unit, t n is the mass density of the nth material component and is expressed in grams per square centimeter (g / cm 2 ) is expressed in units, and S is the response function of the detector. Two possible ways to obtain spectral information are to change the incident photon flux and energy spectrum Φ of the light source at the incident energy level En 0n (E), or a detector with energy spectrum En sensitivity.

[0261] Due to scatter interference issues, spectral imaging has not been implemented in 2D radiographic formats. Scatter removal, as described in, for example, the dual-energy approach in Chao's publication, along with improved database and calibration methods, has now been extended to multi-energy systems. As an example of implementing a spectral imaging system on a 2D detector, such as a system based on a 2D radiographic flat panel, the following is an example of a spectral decomposition method, a three-energy decomposition method.

[0262] Three-energy decomposition method

[0263] The object consists of three material components, for example, molecularly labeled tissue or cells, bone, and unlabeled soft tissue, none of which can be ignored. The exact relationship between the experimental data and the quantities to be found can be expressed as a three-energy X-ray imaging equation system consisting of three nonlinear simultaneous equations:

[0264]

[0265]

[0266]

[0267] Among them, Φ 0H (E), Φ 0M (E), Φ 0L(E) are the energy spectra of the X-ray source at high energy level H, medium energy level M, and low energy level L, respectively. The projected mass density p(x, y) of the first component in the region of interest, the projected mass density b(x, y) of the second component in the region of interest, and the projected mass density s(x, y) of the third component are the total amount of material along the X-ray projection line of the object, expressed in grams per square centimeter (g / cm 2 For example, p is used for plaster casts, which in some cases are mixed with contrast agents such as iodinated agents, b is used for bone, and s is used for soft tissue; μ p (E) is the mass attenuation coefficient of the gypsum casting, μ b (E) is the mass attenuation coefficient of the bone, and μ s (E) is the mass attenuation coefficient of soft tissue. Mass attenuation coefficient μ p (E), μ b (E) and μ s (E) are expressed in square centimeters / grams (cm 2 / g). All these values ​​are known, experimentally determined and well documented. The term [Φ0(E)×exp(-(μ p (E)×p(x, y)+μ b (E)×b(x,y)+μ s (E) × s(x, y))] is the energy spectrum of the X-ray spectrum incident on the detector after passing through the object. S f (E) is the X-ray spectral sensitivity of the detector, which is the amplitude of the electrical signal from the detector as a function of the number of X-rays with energy E after passing through the image object.

[0268] In a three-energy system, the system of equations has three independent parameters: the density of the first component p(x, y), the density of the second component b(x, y), and the density of the third component s(x, y) in the region of interest. The data processing method is to solve the three simultaneous equations and obtain the three measured quantities D H (x, y), D M (x, y) and D L The three unknown quantities p(x, y), b(x, y) and s(x, y) are found in (x, y). The measured quantity D is determined by irradiating the object with X-rays of energy levels H, M and L respectively and acquiring each image of the X-ray detector. H (x, y), D M (x, y) and D L (x, y), such as Figure 9 shown.

[0269] When the object includes more than three components, more X-ray energy levels can be used. Equation (39) is expanded by naturally adding equations for each additional energy level and adding the appropriate μ(E) term to the equation for each additional component.

[0270] There are many methods for solving equation (39). The present invention does not limit the choice of method for solving the equation. Two methods are described below.

[0271] (1) Linearization method

[0272] When the image object is not too thick, the nonlinear absorption effect of the object can be ignored, and the nonlinear equation (39) can be linearized to give a simple linear simultaneous equation:

[0273]

[0274]

[0275]

[0276] Where Ln() represents the natural logarithm of the quantity in the brackets, D H (x, y), D M (x, y) and D L (x, y) represents the signal intensity at pixel (x, y) at high, medium, and low energy, respectively, H0 , Φ M0 , and Φ L0 denote the incident X-ray intensities at high, medium, and low energies, respectively. Since the three left-hand quantities are the measured raw image data (on a pixel-by-pixel basis), the three unknown right-hand quantities p(x, y), b(x, y), and s(x, y) can be found by any standard method of solving three simultaneous linear equations for each point (x, y) on a pixel-by-pixel basis.

[0277] Linearization methods are easy to implement. In many cases, if the goal is to obtain visually distinct images, accuracy should be sufficient. When higher-accuracy quantitative results are required, further correction for nonlinear effects can be performed. The nonlinear correction method for triple-energy X-ray imaging is the same as that for dual-energy imaging and is a well-documented standard procedure.

[0278] Expressed in the more general form of equation (40),

[0279]

[0280] Here, the system of equations is solved by linearizing the n-equation system into a system of n equations, and the material composition c is solved. In a three-energy system, n = 3, and in a four-energy or five-energy to n-energy system, n = 4, 5, ... n. Material decomposition of four or more components can be achieved as described without further modification or significant improvement.

[0281] (2) Multi-step dual energy decomposition method

[0282] Due to the specific behavior of the X-ray absorption coefficient as a function of X-ray energy, for the present invention, a single triple-energy X-ray imaging operation can be considered as consisting of two dual-energy imaging operations: the first step is to treat the image data pairs of the high and medium energy levels as one dual-energy decomposition process, and the second step is to treat the image data pairs of the medium and low energy levels as another dual-energy decomposition process. The theoretical basis of this approach is that at high energy levels, the interaction of X-rays with matter is essentially Compton scattering; at medium energy levels, it is a mixture of Compton scattering and a small amount of photoelectric absorption; and at low energy levels, it is mainly nonlinear photoelectric absorption. Therefore, when the two energy levels are appropriately selected within a sufficiently narrow energy range, the rules for dual-energy decomposition hold.

[0283] In the first step, the X-ray absorption coefficient of the molecular marker tissue or plaster cast material is assumed to be the same as that of bone, allowing the two material systems to be decomposed using a dual-energy method. Consequently, the dual-energy decomposition method can separate the material composition of human soft tissue on the one hand, and the material composition of bone and plaster cast on the other. Then, in a further second step, using dual-energy image data pairs at medium and low energy levels, the material composition of bone and molecular marker or plaster cast can be decomposed.

[0284] One of the advantages of considering the triple energy X-ray imaging decomposition as two separate dual energy decompositions is that all currently available results and methods for dual energy imaging can be directly utilized.

[0285] For multi-energy X-ray imaging decompositions with four or more energy levels, a similar dual-energy decomposition procedure can be employed, for example, obtaining an image of a single component at one energy level (in some cases, at the absorption edge level), and the remaining components at different energy levels, and then iterating the dual-energy decomposition on the remaining components one or more times until a single image for each component is obtained.

[0286] Thus, the methods of the present invention can provide spatial and temporal functional analysis of contrast-labeled materials in a subject, as well as their location and characteristics in space and time, and their location relative to an organ or organ system.

[0287] Alternatively, a multi-energy system can be implemented as a single pulse or a repeating unit of more than one pulse, each repeating unit having a different energy level. For example, a single pulse or a repeating unit of more than one pulse can be used to irradiate an object, where each pulse is designed to have each of multiple energy levels emitted at a different time interval within the pulse. The detector used is energy-sensitive or photon-counting, which can sample at different energy levels at different times within the pulse time interval.

[0288] AI and deep machine learning and artificial neural networks

[0289] One or more criteria can be set by the user or a digital program to search the above database in the material library to generate a component image or multiple component images. Only a subset of the database is searched at a time for decomposition. This search process can be iterated repeatedly, each time using the same or a different (sometimes much smaller) set of databases. This process utilizes methods known to those skilled in the art of artificial intelligence, artificial neural networks, deep neural networks, and convolutional neural networks.

[0290] K-edge method

[0291] The present invention includes a multi-energy system, or alternatively referred to as a spectral imaging system method, which may include the above-mentioned energy decomposition method in combination with a K-edge method. For example, a multi-energy system having n energy levels can be expanded by adopting a K-edge subtraction imaging method, where n is equal to or greater than 3. For example, a component in a region of interest has an absorption edge that is different from other components contained in a background image in the region of interest. The K-edge subtraction method uses a narrowband X-ray spectrum with energies infinitely lower and higher than the K-edge energy of the component. For example, in order to distinguish between multiple contrast-marked components, such as diseased tissue and anatomical markers, the K-edge method can be used together with a dual-energy or multi-energy decomposition method to distinguish or visualize each contrast agent in a background of multiple or overlapping components or tissues.

[0292] For example, in lung imaging, as described in the previous section, microcalcification imaging can be combined with K-edge imaging, where a contrast agent is labeled with a tumor marker. Additionally, K-edge imaging can be applied to the K-edge of the contrast agent to further characterize the tumor region.

[0293] The present invention further expands the method of three or more energy levels combined with the above-mentioned method in DRC component imaging and K-edge imaging, and can be applied to cardiovascular imaging, wherein the multi-energy system separates bone tissue, cardiac tissue, blood vessels, other soft tissues and contrast-labeled tissues. Additional components such as heart valves, stents, surgical tools, catheters and biopsy needles can be distinguished. In spinal surgery, spinal and skeletal tissue can be separated from soft tissue, labeled blood vessels and labeled neural tissue. Using methods familiar to those skilled in the art, the image of the surgical tool can be separated from the background and accurately positioned in 2D or 3D space.

[0294] Interferograms, phase contrast imaging, coherent and partially coherent X-ray imaging

[0295] The present invention also includes embodiments in which interferogram methods familiar to those skilled in the art can be combined with multi-energy decomposition methods to measure at each energy level to improve the discrimination between materials or compositions that otherwise appear similar (e.g., low atomic Z materials).

[0296] Summary of mathematical expressions for multi-energy material decomposition with scatter removal

[0297] Multi-energy system material decomposition can be accomplished using a linear system of equations approach or by iterative dual-energy material decomposition. For the latter, all previous methods and results for dual-energy material differentiation and decomposition, as well as scatter removal methods, can be utilized. Furthermore, through improved databases, simulations, and data synthesis methods, the present invention provides multi-energy systems and methods for rapid and accurate material decomposition, imaging, and quantitative analysis in medical, life science, and nondestructive testing applications. Thus, in general, spectral imaging, through the ability to remove noise and scatter, can now be applied to 2D detector-based radiography.

[0298] Thus there has been shown and described a method for X-ray imaging of an object. As certain changes could be made in the disclosure without departing from the scope of the invention, it is intended that all matter contained in the above description and shown in the accompanying drawings shall be interpreted as illustrative and not in a limiting sense.

[0299] The present invention relates generally to digital X-ray imaging and, more particularly, to a method for digitally imaging an organ or organ tissue or a portion of an organ system using a dual energy device and to a method for separating a single organ X-ray image into component images, each component image representing a single physical substance.

[0300] The method of the present invention utilizes the basic methods for scatter removal and dual-energy X-ray imaging to first separate a mixed chest image into four basic image components: a scatter image, a fat-free tissue image, a fat tissue image, and a microcalcification image. "Microcalcification" is used interchangeably with "calcification." These images are a first-order approximation. The three material compositions of the human chest are then considered. In a second-order approximation, the microcalcification image, the fat-free tissue image, and the fat tissue image are separated so that each contains only a single chest component.

[0301] The method for obtaining a first-order approximation includes the following steps: (a) performing calibration as described below to obtain a pair of numerical relationships for the front detector at high and low energies for microcalcifications c and soft tissue s, thereby obtaining functions c=c(DH,DL) and s=s(DH,DL), (b) performing calibration to obtain a pair of numerical relationships for the front detector at high and low energies for fat tissue and fat-free tissue, thereby obtaining functions f=f(DH,DL) and g=g(DH,DL), (c) irradiating the object with X-rays of average energy level H and average energy L, (d) acquiring high-resolution images DfHh(x,y) and DfLh(x,y) from a front detection position (x,y), wherein the images include primary X-rays and scattered X-rays, and (e) calculating a pair of high-resolution scattered X-ray images DfSHh(x,y) and DfSLh(x,y), (f) calculate a pair of high-resolution primary x-ray images DfPHh(x,y)=DfHh(x,y)-DfSHh(x,y) and DfPLh(x,y)=DfLh(x,y)-DfSLh(x,y), (g) use functions c=c(DH,DL) and s=s(DH,DL) to perform dual energy decomposition on the image pair DfPHh(x,y) and DfPLh(x,y) to obtain two first-order approximate material composition images c1(x,y) and s1(x,y), and (h) use functions f=f(DH,DL) and g=g(DH,DL) to perform dual energy decomposition on the image pair DfPHh(x,y) and DfPLh(x,y) to obtain two first-order approximate material composition images f1(x,y) and g1(x,y).

[0302] The method for obtaining a second-order approximation corrects for microcalcification effects and comprises the following steps: (a) identifying all microcalcification points c1(xk,yk) and all non-microcalcification points c1(xi,yi) in the image c1(x,y), (b) constructing a background image B(x,y) in which the point B(xi,yi)=c1(xi,yi) and in which the point B(xk,yk) is interpolated from the points c1(xj,yj) around the point c1(xk,yk), (c) subtracting the background image B(x,y) from the image c1(x,y) to obtain a second-order approximation microcalcification image c2(x,y), (d) identifying the image c2( (x, y) all zero points c2(x0, y0) and non-zero points c2(xm, ym) in the image; (e) construct a second-order approximate fat tissue image f2(x, y), where point f2(x0, y0) = f1(x0, y0) and where point f2(xm, ym) is interpolated from points f1(xn, yn) around point f1(xm, ym); and (f) construct a second-order approximate fat-free tissue image g2(x, y), where point g2(x0, y0) = g1(x0, y0) and where point g2(xm, ym) is interpolated from points g1(xn,yn) around point g1(xm, ym).

[0303] Another object of the present invention is to provide an improved dual energy calibration method so that the two materials decomposed are actual breast tissue components rather than just equivalent materials. Currently, dual energy calibration is performed by measuring the x-ray attenuation curve using a pair of attenuation plates of different materials with known thickness values. Generally, the materials used for calibration cannot be the same as the materials actually present in the human body because the actual substances in the human body are too complex to be made into quantitative materials. For example, it is a common practice to use aluminum as a representative of human bone material and artificial resin as a representative of average human soft tissue. Therefore, based on the x-ray attenuation in the human body, the dual energy results can only provide an equivalent amount of aluminum and an equivalent amount of artificial resin. The second-order approximation of the present invention cannot be achieved by using standard calibration methods and equivalent decomposition. Therefore, the improved dual energy calibration method is part of the present invention, which is used to decompose the human chest into pure single component images with high precision.

[0304] The calibration method of the present invention comprises the following steps: (1) determining the dual energy equations for two known materials u and v as D H =D DH [u,v],D L =D DL [u, v], (2) Use the standard least squares data fitting method to perform functional decomposition of the energy-related attenuation coefficient function μ u (E)=u p ×μ p (E)+u q ×μ q (E) and μv (E) = v p ×μ p (E)+v q ×μ q (E), to obtain the constant u p 、u q 、v p and v q ; (3) For each coordinate pair (u, v) calculate p = u × (u p +v p ) and q=v×(u q +v q ), to obtain the dual energy equations D H =D DH [p,q],D L =D DL [p, q]; and (4) solving the material p and q as a variable pair (D H , D L ) is a function of the equation system D H =D DH [p,q],D L =D DL [p, q], to obtain the equation system p=p(D H , D L ) and q=q(D H , D L ). This calibration is performed for the component pairs microcalcifications and soft tissue (which is an average combination of fat tissue and fat-free tissue), and then for the component pairs fat tissue and fat-free tissue.

[0305] A multi-energy x-ray source is used to generate a single image for each component in the object using the multi-energy decomposition method of 2D images described in the present invention.

[0306] The prior art describes 3D image acquisition steps for multi-energy 3D image reconstruction by non-rotational CT methods.

[0307] Furthermore, this method for separating and reconstructing images of different atomic z materials, such as microcalcifications or calcifications and soft tissue composed of fat or lipid-dense tissue and fat-free tissue, or contrast-agent-labeled tissue or cells or foreign matter (e.g., microorganisms or inorganic objects or transplanted tissue or stem cells), can be used to construct multidimensional images or imaging sequences in space and time by taking two or more x-ray images in three-dimensional space, for example, using different detectors at different locations or x-ray sources at different locations, or simply moving the same x-ray source or detector to different locations, or moving the object in 3D space, or taking second or more images at different times. When x-ray images of a single material having different atomic z or labeled with different atomic z contrast agents in the object are derived from x-ray images of the object taken at different locations in 3D space, the resulting images of the same material can be combined to provide more information about the location and characteristics of the object and its individual constituent materials or contrast-agent-labeled objects in 3D space. When x-ray images are taken at varying times, for example, sequential images of the object are taken each time, images of the organ 10 and its individual composite materials and contrast-agent-labeled objects are derived. For example, by tracking the dynamic position of selected points on an organ and its components and a labeled object, information related to the position, location, or function, or properties based on movement and motion, of each individual composite material with varying atomic z-numbers within a contrast-labeled object or organ can be compared with information about the organ or organ system. For example, tracking the movement of transplanted stem cells or immune cells, or circulating tumor cells or tissue tumor cells, the migration and characteristics of tumors in 3D space, and the dynamic behavior of such materials or the dynamic behavior of the material or object interacting with the organ can be obtained. Thus, this method allows for functional analysis that includes comparing the motion and properties of a labeled material or object in space and time, its position and properties in space, and its time, position, and conformation relative to the organ or organ system.

[0308] An extension of the dual energy system can be an extension of a single energy system using a k-edge subtraction imaging method. If the goal is to visualize and separate and visualize a contrast agent or calcification or an organic or inorganic component or a mixture of both, which has a different atomic z compared to the background. The K-edge subtraction method uses a narrowband x-ray spectrum with energies that are extremely small below and above the K-edge energy of the contrast material. The A-space method uses a wide spectrum x-ray tube source and uses a photon counting detector to measure the transmission spectrum by pulse height analysis. A further extension of the present invention utilizes three or more energy methods in combination with the above methods. For example, in cardiovascular imaging, a triple energy system separates bone, soft tissue and contrast-labeled cardiac images, while even more energy levels can be used to distinguish a fourth component, such as an implant, such as a heart valve, or a stent or catheter, or a surgical tool from the rest. Or to limit radiation levels, the above method for separating microcalcifications as described is used to further separate the fourth component from the soft tissue and labeled cardiac tissue. The location of the fourth component relative to the background, bone soft tissue, and cardiac tissue can be precisely determined, in some cases to within the μm range in 2D or multiple dimensions, depending on the detector used. Another example is the use of surgical tools or radiation therapy to remove cancerous tumors. The precise location of labeled cancerous and diseased tissue can be located relative to the background. In spinal surgery, spinal and bone tissue can be separated from soft tissue, labeled blood vessels, and labeled neural tissue, and the image of the surgical tool can be separated from the background and precisely located in 2D or 3D space. In automated X-ray inspection, the same method can be used to separate multiple components. And in the characterization of materials, a similar method is used. In baggage scanning and inspection, the same method is used to separate, identify, and locate known and unknown materials and substances.

[0309] The above claims 1, 2, 3, 4, 5, 6, 7, and 8 can be applied to a second x-ray image or further x-ray images 34 taken of an object having overlapping third components, such as microcalcifications or calcifications or atomic z-shaped different materials or materials marked with contrast agents, or foreign bodies. The images can be acquired with different positions of the x-ray source or second x-ray source, or different detectors or second detectors, or simply with the object in a different position. The combined image provides more information about the position or location of the components relative to each other, particularly the third material, which overlaps with a background comprising the first and second or more materials, the images and density information of which can be separated by the multi-energy x-ray system.

[0310] Relative information between all materials can be imaged over time to analyze the position and orientation of each material relative to the others. When acquiring a 2D image, the relative position and orientation of each component can be derived. When a multidimensional image is available, the relative position and orientation of each component can be derived in three-dimensional space.

[0311] 10. As described above, claims 1-8 may also be applied to acquiring two or more x-ray images of an object at different times, such as acquiring an image of the object and each component at a time when acquiring consecutive images of the object. For example, by tracking the position of a selected component in the object, information about the position, location, or dynamic movement and motion characteristics of the selected component in the object can be obtained.

[0312] 11. The aforementioned claims 9 and 10 are combined to provide dynamic and static characterization in 3D and time. Relative position, spatial characteristics, dynamic motion, and component interaction dynamics can be recorded and tracked. For example, this can be used to track stem cells, surgical tools, implants, organic objects or components within organic objects, or mixtures of inorganic and organic objects.

[0313] 12. The preceding claims 1-11 may be applied to systems using triple or more energies, such as spectral energy x-ray sources. In this case, any two energies of the triple or spectral energy x-ray source may be used for imaging, quantification, and image separation, as described for selected composite materials in the object.

[0314] 13. The preceding claim is a fixed or portable system having a power socket.

[0315] 14. The portable system of the preceding claim is based on a battery operated X-ray source and detector assembly.

[0316] 15. Battery operated portable systems based on published patents and provisional patent applications as described in US Patents 5,648,997 and 5,771,269, US Patents 6134297A and 6173034B1, US Provisional Patent Applications 62 / 620,158 and 62 / 628,370 and 62 / 628,351.

[0317] 16. A battery operated or socket powered system that can be folded into a container that can be carried on a luggage bag or rolled up in a luggage roller or in a luggage case.

[0318] 17. A battery operated system having an x-ray input beam management system for the safety of an on-site operator and / or an x-ray shielding system for the safety of an on-site operator and / or a patient in the case of medical purposes.

[0319] 18. A 3D portable system based on all of the above devices and methods by combining two or more 2D images of an object acquired by the above systems, where the x-ray source and the object are moved relative to each other, or set up as a conventional 3D system, where both the x-ray source and the detector are moved, to resolve unknown pixels in the third axis.

[0320] 19. The X-ray source used in the preceding claim is a monochromatic source.

[0321] 20. The aforementioned apparatus is used together with a k-edge subtraction imaging method.

[0322] 21. The aforementioned apparatus and method is an A-space method, wherein the detector is energy sensitive.

[0323] 22. Combine the above methods in 1-19 with K-edge subtraction imaging or A-space method or both to further differentiate materials.

[0324] 23. The above method in 1-22 is used to image inorganic matter having multiple substances.

[0325] 24. The above method in 1-22 is used to image a subject containing organic and inorganic materials.

[0326] 25. Portable, transportable, foldable 2D or 3D systems based on scattering and primary separation methods and dual-energy and multi-energy and spectral energy X-ray sources are described in U.S. Patents 5,648,997 and 5,771,269, U.S. Patent Nos. 6,134,297A and 6,173,034B1.

[0327] 26. A portable, carry-on, foldable 2D or 3D system as described in US Provisional Patent Applications Nos. 62 / 620,158, 62 / 628,370 and 62 / 628,351.

[0328] 1. First, the x-ray source described in the patent emits x-rays of two different energy spectra or a single energy spectrum, while in the present invention, in the calibration step of multi-energy x-ray imaging, in certain applications, the x-ray source emits x-rays of more than two energy spectra.

[0329] 2. The second difference is that in addition to calibrating the main signal of the front detector and the main signal of the rear detector by different atomic z and thicknesses of various components in, for example, in in vivo or ex vivo imaging, not only bone and soft tissue thickness, but also additional tissues or foreign bodies or components are used for calibration, such as surgical tools or implants or contrast markers or third components in the imaging object.

[0330] 3. A third difference is that, during the calibration step, for each energy level image received at the detector, microstructures of varying spatial complexity, size, and compositional complexity are introduced, similar to those expected in the imaged object, that can perturb the x-ray energy spectrum differently. This is used to correlate the primary x-ray signal on the front detector with the primary x-ray signal on the rear x-ray detector at the x-ray wavelength and energy level specified in the system. In some cases, the same method is used to correlate the scattered x-rays from the front and rear detectors. Simplified versions of these microstructures, in terms of complexity and composition, may be used.

[0331] 4. The fourth difference is that in one embodiment the rear detector is replaced by a single or multiple detector units at each position of the selector material of the beam selector.

[0332] 5. The beam selector can be translated or moved in three dimensions, or the focus can be adjusted manually or automatically using actuators and electronic controls to allow flexibility in the x-ray emission position of the x-ray source.

[0333] The term "selected location" is defined as a location on the x-ray-sensitive medium of the rear detector 26 where, due to the function of the beam selector, only primary x-rays are received, and x-rays scattered from this location are substantially blocked. A "selected projection line" is defined as a straight line connecting the x-ray source 12 to a point in the "selected location." Typically, this point is near the center of the selected location. Note that for the rear detector assembly 26 of this embodiment, only the signal at the selected location is utilized. The rear detector cell at the selected location has a fixed geometric relationship with some of the front detector cells. This relationship is established by drawing a selected projection line from the x-ray source 14 via the beam selector 18 to the selected location. The selected projection line intersects the rear detector surface at the rear detector cell at coordinate (i, j) and intersects the front detector, having a fixed geometric relationship with some of the front detector cells. This relationship is established by drawing a selected projection line from the x-ray source 14 via the beam selector 18 to the selected location. The selected projection line intersects the rear detector surface at the rear detector element at coordinates (i, j) and intersects the front detector surface at the front detector element at coordinates (x(i), y(j)). Here, (x(i), y(j)) represents the Cartesian coordinates (x, y) of the front detector element in the front detector assembly 16 closest to the selected projection line. The image file Dr1(i, j) acquired from the rear detector assembly 26 contains only the signal at the selected location where the primary x-ray is received, and scattered x-rays are substantially blocked. The data at image pixel (i, j) is obtained from a single detector element or from a combination of a small number of detector elements around the selected projection line. Similarly, Dfl(x(i), y(j)) represents an image file with low spatial resolution from the front detector assembly 26. The data at image pixel (x(i), y(j)) is the data from a single detector element or from a combination of a small number of detector elements around the selected projection line. The relationship between (i, j) and (x(i), y(j)) is experimentally established and stored for all apertures 20 of the beam selector 18.

[0334] The fourth difference is that in one embodiment, the rear detector is replaced by a single or multiple detector units at each position of the selector material of the beam selector as described in provisional patent applications US62677312 and US62645163

[0335] The beam selector can be translated or moved in three dimensions, or the focus can be adjusted manually or automatically by actuators and electronic controls to allow flexibility in the x-ray emission position of the x-ray source, as described in U.S. Provisional Patent Applications 62 / 620,158, 62 / 628,370, 62 / 628,351, 62677312, and 62645163.

[0336] There are two preferred embodiments for using a beam selector and a single energy or dual or multi-energy x-ray source. Both device embodiments include an x-ray source, a two-dimensional front detector, a beam selector, and a two-dimensional rear detector. Typically, the beam selector passes some x-rays to the rear detector and blocks other x-rays from the rear detector. The difference between the device embodiments is that the beam selector passes and blocks x-rays. In a first embodiment, the beam selector passes only primary x-rays to the rear detector and blocks scattered x-rays. In a second embodiment, the beam selector passes only scattered x-rays to some locations of the rear detector, blocks primary x-rays to these locations, and passes primary x-rays and scattered x-rays to the remaining locations of the rear detector.

[0337] Both method embodiments include the following steps: (a) irradiating an object with x-rays from an x-ray source, (b) generating a low-resolution primary x-ray image at a rear detector DrP1, (c) calculating a low-resolution primary image DfP1 at a front detector along a selected projection line, (d) generating a high-resolution image Dfh from the front detector, (e) generating a low-resolution image from Dfh at the front detector Dfl, (f) subtracting DfP1 from Dfh to determine a low-resolution scatter component DfSl, (g) smoothing the low-resolution scatter component DfSl by removing high spatial frequency components, (h) calculating a high-resolution scatter image DfSh by interpolating the smoothed low-resolution scatter component DfSl, and subtracting the high-resolution scatter image DfSh from the high-resolution image Dfh to obtain a high-resolution primary x-ray image DfPh.

[0338] Another preferred embodiment is the first and real-time X-ray measurement of single energy, dual energy, triple energy based on the need for material decomposition and, in some cases, based on scatter removal of the target or region of interest, or quantitative analysis of the composition of unlabeled or contrast agent labeled areas, each energy being generated by one pulse, or two pulses, or three pulses or more pulses with one or two or more energy distributions.

[0339] Apparatus for a closed-loop feedback system for reducing x-ray dose

[0340] In a preferred embodiment, the x-ray beam radiation output level on the region of interest is adjusted based on the first image or the first group of first images acquired for the target in the region of interest, or the first image of the region of interest or the first images at dual energy levels or multiple energy levels, and the x-ray beam is spatially adjusted to irradiate only the region of interest or the target, thereby minimizing the input x-ray dose for subsequent first measurements and live measurements, without including data acquired for visualization and quantitative analysis.

[0341] Figure 8A preferred embodiment is shown in which the collimator 10 202 has transmissive regions 200 alternating with opaque regions 201 .

[0342] exist Figure 8 In FIG. 2 , an embodiment of a collimator 202 has transmissive regions 200 alternating with opaque regions 201 .

[0343] The x-ray beam from the x-ray source can scan in a pre-programmed pattern during one or more frames of x-ray sampling. Alternatively, the x-ray source can simply irradiate the entire region 202 or a selective region of 202 to generate or selectively generate x-ray nanobeams.

[0344] Improved kit configuration

[0345] In a preferred embodiment, the present invention also includes retrofitting hardware components and software to modify existing hardware and software that a user may already have, such as an X-ray source capable of generating X-rays at energy levels relevant to a particular application.

[0346] · Medical world, such as 20KeV-1000KeV · Radiation therapy, Mkev

[0347] Monochromatic sources, either of arbitrary energy levels, or 0-70 keV, such as used in synchrotrons and similar sources. Alternatively, such sources can have higher energy levels.

[0348] • Monochromatic sources obtained from conventional X-ray tubes can be any of the X-ray tube energy levels produced by filtering and tailoring of the anode target.

[0349] Ultrafast X-ray source

[0350] The modification kit may include any one or more of the following:

[0351] 1. Calibration kit including hardware and software

[0352] 2. Software for calibrating the method of the present invention

[0353] 3. One or more collimators to modify the output from the x-ray source beam for scatter removal or material resolution imaging.

[0354] 4. Hardware and software for modifying the x-ray source and x-ray source control to switch from different energies

[0355] 5. Add one or more X-ray sources as described in the present invention

[0356] 6. New X-ray sources replace existing X-ray sources

[0357] 7. The X-ray detector assembly as described in the scatter removal and material decomposition device of the present invention, instead of the membrane

[0358] 8. Software for imaging process and / or acquisition

[0359] 9. Hardware positioning or moving device for moving the x-ray source or other parts of the x-ray system involved in the method of the present invention

[0360] 10. Beam selector for modifying existing double or multi-layer detectors

[0361] 11. If a detector is already present, the beam selector plus the detector completes the dual detector scatter removal assembly

[0362] 12. One or more detectors where a beam selector or collimator and detector are already present.

[0363] 13. Tunable hardware, such as MEMs or crystals, for beam steering or adjusting the field of view and other output characteristics of the x-ray beam or selecting nanobeams

[0364] 14. X-ray beam position steering device or electron beam steering device

[0365] 15. If an x-ray source already exists, add one or more x-ray sources or hardware to create more x-ray emission locations.

[0366] 16. Any additional hardware required for spectroscopic absorptiometry or X-ray microscopy

[0367] 17. Other hardware required for spectroscopic or optical analysis systems including x-ray or non-x-ray imaging modalities and techniques, and including x-ray spectroscopic absorptiometry, or x-ray microscopy, spectroscopy, MRI, PET, optical averaging, photoacoustic, ultrasound, thermal imaging and analysis.

[0368] Various contrast agents are modified and linked to each other to achieve co-localization sensitivity of two or more imaging modalities or imaging methods, such as photoacoustic imaging or PET or MRI, or optical coherence tomography, or bioluminescence or fluorescence imaging or ultrasound imaging. Each form of contrast agent can be chemically linked to ensure co-localization.

[0369] Additionally, further modifications such as micelles or nanomicelles or lipidated forms of the molecules or any combination thereof are used.

[0370] The present invention includes contrast agents that can be used in a variety of applications with flat-panel and rotational MRI systems, as well as ultrasound, and contrast agents for fluorescence and optical methods such as microscopy, endoscopy, and photoacoustics.

[0371] Other aspects of the present invention are as follows:

[0372] Naturally occurring or non-toxic reagents include air, gases such as intrabone gas and intradiscal gas, air spaces in the lungs, or cation-rich regions present in arthritis, and other examples of any X-ray detectable region that can be distinguished from the rest of the region of interest, including those produced by enzymatic activity, including molecular aggregates in regions such as intracellular regions.

[0373] The present invention encompasses in vivo liquid biopsies.

[0374] The present invention encompasses measurement of physiological states such as oxygenation state, changes in state such as motion or oxygenation or previously measurable events that trigger changes in state in vivo by optical methods, spectroscopic methods, molecular interactions, in vivo flow dynamics and flow rates, which can be measured by 2D or 3D x-ray quantification methods as described herein.

[0375] The invention includes measurements of molecules, atoms, cells and structures, or phenotypes or motion or fluid dynamics, which can be triggered by internal or endogenous chemical, electrical, electromagnetic, electrochemical, mechanical, acoustic events, magnetic or a combination of two or more.

[0376] The present invention includes measurements of molecules, atoms, cells and structures, or phenotypes or motion or fluid dynamics, which can be triggered by external forces due to interaction with a target or region of interest through chemical, electrical, electromagnetic, mechanical, electrochemical, magnetic, acoustic or a combination of two or more of these external force-based events combined with internal events.

[0377] For example, fast events that characterize the dynamics of atoms and molecules, or nanostructures, microstructures, cells, or a combination of one or more events

[0378] Contrast agent levels

[0379] For organic and non-organic objects with different atomic Zs from each other, one can simply mix different atomic Z materials or radioactive labels (such as iodine) with the substance to be imaged to achieve the required density for visualization in 2 dimensions.

[0380] The proportion of radiolabel required to visualize these materials will need to be, for example, 1) allow the bone cast to solidify and achieve the time-varying stiffness and stability required for bone healing to occur and other intended functions of the cast; and 2) allow quantification and visualization in X-ray imaging and therefore separate the image of the cast from images of human organs / tissues such as bone or soft tissue.

[0381] To achieve 2), the following formula needs to be considered to estimate the required density of x-ray detectors in the mixture to sense the signals required for imaging and quantification.

[0382] The X-ray transparency of a substance depends primarily on its density. Theoretical and experimental studies have shown that when an X-ray beam traverses a medium, the intensity of the beam decreases due to the absorption and deflection of photons by the medium. The degree of X-ray attenuation follows the following formula:

[0383] I=Ioe-μx

[0384] Where I is the transmitted beam intensity, Io is the incident beam intensity, and x is the thickness of the medium. The mass attenuation coefficient μ is expressed as μ = ρZ4 / AE3.

[0385] Where ρ is the density, Z is the atomic number, A is the atomic mass, and E is the x-ray energy. Therefore, low energy x-rays and high atomic number materials have high x-ray attenuation.

[0386] Therefore, based on this formula, for example, in bone cast materials or battery materials or microchip materials, two or more two-dimensional images can be further expanded to form two-dimensional layered images, or 3D images, quantitative imaging data and differentiated material quantitative data, and the density measurement value of such material can be derived from the density measurement value of the 2D data.

[0387] Another example is bone cement or casting material or biofilm. The cement or casting material is mixed with a contrast agent (e.g., an iodinated or other atomic z-variant labeled molecule or derivative thereof) to achieve the desired radiodensity for X-ray detection. Alternatively, inorganic compounds, such as iron sulfate, silver-coated microparticles or 1-chloronaphthalene, helium, hafnium, or even nanoparticles, are other contrast agents used for in vivo imaging.

[0388] III. Methods for Mixed Labels

[0389] In one embodiment, the material to be imaged is mixed with a marker that can be identified by X-ray or hybrid imaging modalities.

[0390] Example: 1. Gypsum casting

[0391] Mix the contrast medium evenly with the plaster cast. The contrast medium may be pre-tinted to ensure even mixing by visual inspection.

[0392] Add water

[0393] Fiberglass casting

[0394] Mix the contrast agent homogenously with the resin. The contrast agent can be combined or mixed with the pigment to ensure visual verification of homogenization.

[0395] Adding catalyst to cure glass fiber

[0396] Another aspect of the invention is to use a 2D flat panel detector to correlate image and density measurements and composite analysis within a single material of an object composed of two or more materials.

[0397] The present invention includes methods for analyzing the relative composition, density and image information of regions within a single component and the relative composition, density and image information of some components relative to other components in terms of position, density and image (including morphology) and image size (such as tumor size or disease tissue size).

[0398] One aspect of the present invention is disease diagnosis performed using 2D flat panel quantitative imaging methods and multidimensional methods based on 2D flat panel quantitative imaging methods, which include quantitative aspects of conventional CT scanners, which are typically used in 3D formats. The 2D flat panel quantitative imaging methods use separation of tissue images, analysis of the location, density, and dimensionality of individual tissues and other materials or tissues, as well as motion measurements and relative measurements of components of these parameters. This analysis can be performed independently over time. Examples include cancer diagnosis, circulatory (blood) system diseases and conditions, such as coronary artery disease (atherosclerosis), vascular aneurysms, and blood clots; spinal conditions; kidney and bladder stones; abscesses; inflammatory diseases, such as ulcerative colitis and sinusitis; and injuries to the head, skeletal system, and internal organs. Because the current quantitative and high-resolution images are parallel to CT images, 2D flat panel images can reveal details and quantitative information through scatter removal methods and separated tissue images and quantitative measurements related to size and density and images.

[0399] For example, a pulmonary embolism, or a blood clot in the lungs, may require spiral CT to see the details of various tissues for diagnosis. However, current methods using 2D flat panels require much lower radiation to achieve the detail and quantitative analysis information needed to make a diagnosis.

[0400] This applies to many different types of tissue, including lungs, heart, bones, soft tissue, muscle, and blood vessels.

[0401] Another aspect of the present invention is material characterization and identification in industrial environments where CT scanners are required, a 2D flat panel based system may be sufficient to quantitatively analyze the presence, location, characterization and identification of materials or substances embedded in objects in industrial applications such as cargo inspection, security x-ray and automated x-ray inspection.

[0402] The present invention relates generally to digital X-ray imaging, and more particularly to digital X-ray imaging using a 2D flat panel for combined quantification and image analysis of individual materials in a subject in time and space, with some capabilities intended to replace the use of conventional CT scanners.

[0403] Another embodiment of the present invention is configured with PET or optical imaging, or MRI or ultrasound or acoustic or photoacoustic imaging methods.

[0404] One aspect of the present invention includes X-ray particle image velocimetry to measure flow using particles, such as microbubbles, as tracer particles for studying hemodynamics and circulatory vascular disease. This invention is particularly useful for deep tissue fluid flow measurement, as overlapping tissue and scattering reduce the visibility and quantification capabilities of methods based on 2D X-ray detectors.

[0405] Stereo PIV uses two detectors with independent viewing angles to accurately perform z-axis displacements. Alternatively, 3D acquisition using 2D flat panel-based imaging methods that rapidly acquire multi-dimensional representations can be fast enough to acquire velocities in 3D space.

[0406] Holographic PIV is also part of the present invention using an interferogram based approach to

[0407] II. Methods

[0408] 1. A two-dimensional flat panel based single, dual and triple energy or multi-x-ray imaging system for acquiring one or more two-dimensional images of an object from different positions and / or at different times, the system comprising:

[0409] (a) In a physical sequence from front to back, an x-ray source, a two-dimensional x-ray detector, the object being a body or an object or a region of interest of the object located between the x-ray source and the x-ray detector;

[0410] (b) the x-ray source is adapted to emit x-rays having three different energy spectra to pass through the object;

[0411] (c) the two-dimensional X-ray detector receives X-rays from the X-ray source and converts image information contained in the emitted X-rays into electrical signals and transmits them to a computer;

[0412] 2. A variation of the embodiment is to add a mover to the hardware to move the x-ray source relative to the object so that 2D images can be acquired at different angles and combined.

[0413] 3. Another implementation of the hardware, along with #1, is to add a mover for moving the x-ray source and the detector relative to the object.

[0414] 4. Another embodiment, along with the hardware in #1, is the arbitrary movement of the object if it is a living organism or animal, or internal robotics of the object to move one or more components of the object or the entire object.

[0415] 5. Another embodiment of the present invention is to be configured together with PET or optical imaging, or MRI or ultrasound or acoustic or photoacoustic imaging methods.

[0416] 6. The x-ray imaging system of claim 1, wherein one of the energy spectra has an average energy in a range from approximately 15 keV to 200 keV.

[0417] 7. The x-ray imaging system of claim 1 , wherein the object is a body;

[0418] 4. According to the X-ray imaging system described in #1, the object is an organic material or a mixture of organic and organic materials.

[0419] 8. The x-ray imaging system of claim 1, wherein the apparatus and method described in the patent separate and analyze images of individual components within the object based on size, composition, thickness, shape, morphology, relative position and density relative to the rest of the object, and relative movement and position in time and space.

[0420] a. Apparatus and method for removing scatter from an X-ray image using a two-dimensional detector and a mono-energy spectrum X-ray source, U.S. Patent 6134297

[0421] b. Apparatus and method for dual-energy X-ray imaging, Patent No.: US6052433

[0422] c. Apparatus and method for dual-energy X-ray imaging, Patent No.: US6052433

[0423] d. Apparatus and method for removing scatter from x-ray images, Patent No.: US 5771269

[0424] e. Apparatus and method for removing scatter from x-ray images, Patent No.: US5648997

[0425] f. U.S. Provisional Application #62692675, 3D 3E Calibration; 62620158, 62628370, 62628351, 62677312, 62700157, 62711522, 62697174, 62620158, and 62645163 - Provisional patent applications filed by Zhao, relating to the subject matter of scatter removal, dual-, triple-, and multi-energy, multi-dimensional, molecular imaging, and contrast agents, and methods and 3D X-ray imaging.

[0426] 9. An x-ray imaging system according to #1, wherein images and quantitative measurements of individual components within an object can be separated and analyzed based on parameters such as size, composition, thickness, microstructure, shape, morphology; one or more regions of the same component, their relative position, location and measurements of the aforementioned parameters, compared to the rest of the object, individually and / or in high resolution and real time and / or in time periods and / or in 2D and multi-dimensional space, and their relative position, location and measurements of the aforementioned parameters compared to other one or more regions of the same component, including density, and relative movement, relative position, size, composition, thickness, shape, morphology, microstructure, addition or loss of contents.

[0427] 10. The results can be used to diagnose various diseases, such as the size of vascular features, the presence of clots, irregularities, microcalcifications, special materials or cysts, fractures, increased density in an area, loss of tissue content, addition of tissue fragments, specific microstructures, derivatives of composition and changes caused by density measurements and images, especially in cases where high resolution and accuracy and quantitative measurements are required, for example, CT scanners, bone scanners, MRI and densitometers will have to be used together or separately to obtain the required results, or in some cases, will not be sufficient to give a satisfactory answer for a relatively certain conclusion.

[0428] 11. The results are used for surgical guidance, especially for minimally invasive surgery, radiation therapy and biopsy, especially in cases where CT scanners, bone scanners, MRI and densitometers must be used together or independently.

[0429] 12. The results are used for identification and characterization of composition, materials, substance failure analysis, component inspection, especially where a CT scanner is usually necessary.

[0430] 13. Such a system can be portable

[0431] 14. Such a system can be battery operated.

[0432] 15. Such a system can be portable in a field setting and can be packed into a carrying bag.

[0433] 16. Such a system is portable and can be packaged into a packaging size box.

[0434] 17. Scattering and interference patterns of primary x-rays based on the above hardware, and the addition of diffraction gratings or beam splitters after the x-ray source and before the object, are used for the measurement of the velocity of blood and other biological fluids in the diagnosis of diseases.

[0435] 18. X-ray based particle image velocimetry combined with tissue separation and time measurement for velocity measurement.

[0436] 19. Using the above hardware and methods, the relative density and image of a region of interest within a first component compared to the remaining components of the first component, as well as the relative density and image of different components in regions of interest adjacent to or associated with the region of interest within the first component, can form indicative information for disease diagnosis or characterization or identification of materials or compositions. Timely monitoring of this information is effective in diagnosing diseases at an early stage. Examples of diseases for which a quantitative 2D flat panel x-ray system can be used instead of a CT scanner include:

[0437] In stress fractures, callus formation may occur near the fracture, affecting the bone measurements and the surrounding tissue.

[0438] Compared with normal bone density, the density changes in the injured area are atypical, and the density changes in the unaffected bone area and healthy tissue are uneven.

[0439] Vascular calcification

[0440] The present invention, which uses a scatter-free 2D flat-panel imaging method instead of neuro-CT scans, is used to diagnose and monitor disease conditions and treatment responses of the brain and spine. It detects irregularities in bones and blood vessels, certain brain tumors and cysts, herniated disks, epilepsy, encephalitis, spinal stenosis (narrowing of the spinal canal), blood clots or intracranial bleeding in stroke patients, brain damage from head injuries, and other conditions. Many neurological disorders share certain characteristics, and CT scans can aid in the correct diagnosis by differentiating the area of ​​the brain affected by the disorder. 2D flat-panel-based quantitative imaging, as described in the patent and provisional patent, can now replace this.

[0441] Diagnosis, treatment, and long-term monitoring of pain management

[0442] Muscle disorders

[0443] Bleeding in the brain, pointing to the location of a tumor, infection, or blood clot.

[0444] Guidelines for procedures such as surgery, biopsy, and radiation therapy.

[0445] Localization of suspended cancer cells, stem cells, rare cells, and foreign objects.

[0446] Treatment and surgical planning and guidance as well as treatment and treatment response and post-treatment monitoring of other organs in the body, kidneys, limbs, eyes (implant placement).

[0447] 20. One aspect of the present invention is to enable characterization and identification of materials 20 in industrial settings where a CT scanner is required, a 2D flat panel based system may be sufficient to perform quantitative analysis of the presence, location, characterization and identification of materials or substances embedded in objects in industrial applications such as cargo inspection, security x-ray and automated x-ray inspection.

[0448] 21. One aspect of the present invention is the ability to measure fluid flow characteristics and identification and velocity in 2D or 3D space in industrial environments where conventional CT scanners, optical, acoustic systems or any other available systems and methods are not effective and deterministic.

[0449] The present invention is a compact pulsed X-ray source that provides single-pulse X-ray pulses with an X-ray output corresponding to stored electrical energy between 100 and 1000 joules per pulse and a typical pulse duration between 0.1 and 10 ms. This X-ray source is lightweight, compact, and requires a very low power supply. It is suitable for human imaging.

[0450] (3a) Use a vacuum-sealed field emission tube. This is a cold cathode X-ray tube. The use of a cold cathode X-ray tube can significantly improve energy utilization efficiency. We note that the total electrical energy required to provide the pulsed X-rays used to generate the X-ray image frames is not large. For example, an average electrical energy of about 500 J is sufficient. To obtain 500 J of electrical energy, it is equivalent to turning on a 100 W light bulb for 5 seconds. However, when using a heated cathode X-ray tube, at least 10 times this energy is required because the energy utilization efficiency is low. When using a cold cathode tube, an energy utilization efficiency of 50% to 70% can be achieved.

[0451] (3b) The electrical energy for single-pulse operation is first stored in a capacitor. The capacitor is built up in the previous stage of operation at a much lower voltage. The voltage of the capacitor is selected between 5 kV and 10 kV.

[0452] (3c) The electrical energy stored in the capacitor is delivered to the X-ray tube via a high voltage pulse transformer to provide a voltage pulse of 100 kV to 150 kV.

[0453] (3d) The pulse duration is controlled to be between 0.1 ms and 10 ms. Accordingly, the current flowing in the tube will be between 10 A and 1 A to provide the required amount of energy for quality x-ray imaging. The pulse width and current are determined by the capacitance of the capacitor, the inductance of the pulse transformer, and the parameters of the tube's V-1 characteristic curve. The pulse current flowing in the x-ray tube is substantially lower than that of nanosecond flash x-rays, where the current is between 1,000 A and 100,000 A. The reduced current is very beneficial for enhancing other performance parameters, such as focus size and tube life.

[0454] In one embodiment of the invention, the flash X-ray source is based on the use of a field emission tube driven by a pulse transformer triggered by an HV transistor.

[0455] In summary, the present invention comprises an x-ray source comprising major components with typical parameters, including a 2.5 kV DC power supply, a high-voltage capacitor having a capacitance of 2 μF, an electronic trigger circuit, a high-voltage pulse transformer encapsulated in silicone resin, and an x-ray tube contained in a plastic housing. The total energy stored in the capacitors for generating a single x-ray pulse is 30 J at HV = 50 kV. The entire x-ray source measures approximately 8" x 8" x 16" and weighs approximately 30 pounds.

[0456] 1. Significant reduction in size and weight. Currently, the best X-ray sources with heated cathodes whose output corresponds to 100 J of electrical energy per second (10 mA, 125 kV) are approximately 100 lb. A cold cathode-based pulsed X-ray source with an output corresponding to 100 J of electrical energy per pulse would be only 30 lb, suitable for animal studies. A pulsed X-ray source with 500 joules per pulse would be about 50 lb, suitable for human imaging.

[0457] 2. Substantially reduces power supply. To convert 100 J or 500 J of electrical energy into X-ray pulses, a heated X-ray source typically requires 50 to 100 times more electrical energy to support system operation (for heating the filament, for maintaining the 100 kV high-voltage DC power supply, and for tube cooling). To convert the same amount of energy into X-rays using a field emission tube, the energy required to support system operation is only 1.5 times, or a total of 150 J or 750 J of electrical energy (for charging the HV capacitor at 5 kV to 10 kV). This is very advantageous for portable units that use batteries as a power source.

[0458] A device and method for performing triple-energy X-ray imaging in Case 1 to separate three materials of different atomic z numbers are described. This method is used to visually separate vascular / nervous tissue from bone and other soft tissues under surgical guidance, separate tissue, diseased tissue, or tumors labeled with antibodies bound to particles of varying atomic z numbers, or remove overlap effects caused by the presence of plaster casts in human X-ray imaging. The device physically comprises an X-ray source and a two-dimensional X-ray detector. The object is composed of three or more composites of varying atomic z numbers, such as a human organ containing tissue- or diseased-tissue-specific molecular markers or a human structure overlapped by plaster cast materials for medical purposes. The object is positioned between the X-ray source and the X-ray detector. Using a triple-energy data decomposition method, a three-material composition image is obtained, including a bone mass density image b(x,y), a soft tissue image s(x,y), and a plaster cast mass density image p(x,y) or a molecularly labeled tissue mass density image p(x,y).

[0459] The present invention provides a method and apparatus for separating images of human organs from overlapping effects caused by plaster casts. The image subject is a human organ or structure with some overlapping plaster cast support. The separated image contains complete information for a separated bone image, a separated soft tissue image, and a separated image of the overlapping plaster cast; the last is typically discarded. Thus, the present invention is triple-energy X-ray imaging. The system acquires three X-ray images at three different X-ray energy states and, after subsequent data processing, provides three separate material composition images as described above. Another notable aspect of the present invention is the use of a slightly modified plaster cast material. Currently available plaster casts use standard, well-established materials primarily composed of calcium sulfate (CaSO4) in various hydrate forms. This plaster cast material possesses many excellent chemical, physical, and mechanical properties, making it suitable for developing medical devices for human body support. However, this classic plaster cast material has drawbacks in meeting the emerging demands of digital X-ray imaging. The X-ray absorption coefficient of this classic plaster cast material is too close to that of bone, which is also primarily composed of calcium compounds. An undesirable consequence is that if high-quality separated human images are to be obtained, the corresponding imaging system must be constructed with high precision. To avoid this problem, the present invention advocates the use of a newly created digital-x-ray compliant plaster casting material. The digital-x-ray compliant plaster casting material has all the significant chemical, physical and mechanical properties compared to the standard gypsum casting material, except that its average atomic number Z is very different. The manufacture of the digital-x-ray compliant gypsum casting material is described in U.S. patent application No. ***. The following is a brief description. A typical digital-x-ray compliant gypsum casting material can be a mixture of two materials: mainly conventional gypsum casting material, with a small amount of barium sulfate (BaSO4) added. The present invention does not exclude the use of conventional gypsum casting materials, however, if the new gypsum casting is made of digital-x-ray compliant gypsum casting material, the separation of the gypsum casting and the human image can be performed very effectively and economically.

Claims

1. A method for imaging an object having n≥3 different material compositions t, the method comprising: (a) providing an X-ray source and an X-ray measuring device having at least one 2D detector, wherein the object is located between the X-ray source and the X-ray measuring device; (b) irradiating the object with X-rays having n different energy levels from the X-ray source, the X-rays passing through the object to be received by the X-ray measuring device; (c) obtaining measurements at n ≥ 2 energy levels, wherein at least one measurement is performed for each of the n energy levels; (d) calibrating at each energy level for each material component and each combination of the material component and two or more components and establishing a database using measurements derived without scattering perturbations; (e) Scattering is removed from the measurements at each spectral level to generate an image D for each of the n energy levels 1…n ; (f) Generate n equations D 1…n =∫[Φ 0n (E)×exp(-(μ1(E)×t1+μ2(E)×t2+…μ n (E)×t n )]×S(E)dE, where, Φ 0n (E) is the energy spectrum of the X-ray source at the nth energy level, μ n (E) is the mass attenuation coefficient of the nth material component, t n is the projected mass density of the nth material component, and S is the response function of the detector; and (g) Solving the system of equations to generate at least one image for each of the n material compositions t.

2. The method according to claim 1, wherein By linearizing the n equations into n linear equations L n (D1… n / Φ 0n )=μ1(E)×t1+μ2(E)×t2+…μ n (E)×t n to solve the system of equations and solve for the material composition t.

3. The method according to claim 2, wherein: The n linear equations are solved with reference to a database established by calibration on actual materials and known materials.

4. The method according to claim 1, wherein The system of equations is solved using multiple dual energy decompositions, wherein for at least one material t1 and materials t2 ... t with a maximum number of materials n-1, n The first dual energy decomposition is performed on the composite of the material composition, the material t2, and the material t3 ... t with the maximum material number n-2. n A second dual-energy decomposition is performed on at least one of the compositions, and the process continues until n-1 dual-energy decompositions are performed.

5. The method according to claim 4, wherein The decomposed component data were further quantified using the K-edge method for the contrast-marked components.

6. The method according to claim 4, wherein: The results and intermediate results of the iterative dual-energy decomposition are further decomposed using a dual-energy post-imaging analysis method developed for a prior art dual-energy system, the dual-energy post-imaging analysis method including a unique and rare component decomposition method defined using established linear relationships between known materials, materials of unknown synthetic components, and individual components, and deriving decomposed component images at selected areas where unique and rare components are located through measurement values ​​at dual energy levels or more.

7. The method according to claim 1, wherein The object is irradiated with X-rays at n discrete energy levels.

8. The method according to claim 1, wherein The object is irradiated with broad spectrum X-rays, and the received X-rays are measured using energy sensitive or photon counting methods.

9. The method according to claim 1, wherein: Calibration is performed using each actual material composition and composites of the actual material compositions having various thicknesses.

10. The method according to claim 1, wherein Using a database built from stored radiographic measurements of the same or similar materials, an image is generated based on derived data for each material composition by reference to derived measurements of a single composition comprising one or more physical substances.

11. The method according to claim 1, wherein Using a database of stored CT, MRI, SPECT, spectroscopy, photoacoustic and other energy, chemical or electrochemical measurement modalities, images are generated based on derived data for each material component by referencing derived or synthesized data from a single component comprising one or more physical substances.

12. The method according to any one of claims 9 to 11, wherein: The database includes one or more of the physical, chemical, electrical, physiological and dynamic properties of the material components in static, temporal and spatial form.

13. The method according to claim 12, wherein: A subset of a database of structured or unstructured data is referenced based on one or more criteria set by a user or a digital program.

14. The method according to claim 13, wherein The criteria include one or more of algorithms or methods used in artificial intelligence methods, deep machine learning, artificial neural networks, convolutional neural networks and deep neural networks.

15. The method according to claim 12, wherein: The properties include one or more of characteristic regions of at least one marked region of interest having physical, chemical, spatial, temporal and physiological forms, atomic z number, density, unstructured molecules, structured molecules, synthetic molecules, microstructure, structure.

16. The method according to claim 12, wherein: The material component is one or more tissue regions, contrast-marked plaster casts, materials and cavity forms, regions of one or more organs, molecular complexes, contrast-marked molecular complexes, contrast-marked organelles, ion mixtures, ablated regions of tissue, tumors, diseased regions of tissue, chemical compositions, semiconductor compositions, metallic moieties, inorganic or organic substances, or mixtures of inorganic and organic substances.

17. The method according to claim 1, wherein The X-ray measurement includes one or more measurements of interferogram, phase contrast imaging, coherent X-ray imaging and partial coherence X-ray imaging.

18. A method of imaging an object, the method comprising: providing an X-ray source and an X-ray measurement device having at least one two-dimensional detector, with an object disposed between the X-ray source and the X-ray measurement device, wherein the X-ray measurement device is configured to reduce the influence of X-ray scattering from detected X-rays to generate at least one image, and the X-ray measurement device is configured to perform pre-measurement calibration, the pre-measurement calibration being performed using each material component and a composite of each material component having various thicknesses; irradiating the object using the X-ray source with X-rays having at least two or three different energy levels; detecting the X-rays by the X-ray measuring device after the X-rays have passed through the object; reducing the effects of X-ray scatter from detected X-rays; At least one image is generated.

19. The method of claim 18, further comprising identifying at least one substance in the at least one image by the X-ray measurement device.

20. The method of claim 19, further comprising identifying a plurality of different substances in the image.

21. The method of claim 19, further comprising identifying in the at least one image a substance indicative of an abnormal condition or a diseased state and / or at least a change in a diseased state.

22. The method of claim 19, further comprising identifying the substance in one of its physiological, chemical, physical, temporal and / or dynamic states and / or at least one state of a dynamic process and / or motion.

23. The method according to claim 22, wherein One of its physiological, chemical, physical, temporal and / or dynamic states and / or at least one phase of a dynamic process and / or movement is indicative of an abnormal condition and / or a pathological state and / or at least one change in a pathological state.

24. The method according to claim 23, wherein Identifying the substance also includes determining a state of the substance that is different from a normal or original state due to an energetic, chemical, or mechanical perturbation.

25. The method according to claim 19, wherein The substance is one or more of the following: calcification, microcalcification, biopsy tool, surgical tool, surgical tool, implant, plaster cast, plaster cast mixed with contrast agent, tissue, molecule, cell, cell cluster, molecular complex, cell cluster, chemical substance, material, nanoparticle, or particle of atom z different from the rest of the atoms in the region of interest, or particle derivatives combined with cells, tissue, iodide, barium sulfate, molecular derivatives and molecular markers and / or other existing CT or X-ray markers.

26. The method according to claim 19, wherein The substance is a tumor.

27. The method according to claim 26, wherein The tumor is labeled with at least one contrast agent.

28. The method according to any one of claims 19 to 27, wherein The at least one substance is identified using one or more of a database of known materials, a database of materials identical or similar to the substance, a database of measurement data from known materials for which a quantitative relationship based on the measurements has been established with the unknown material, a database of synthetic data, and / or a database of empirical data.

29. The method according to claim 28, wherein The database is used to derive a quantitative relationship between the physical properties, thicknesses and / or characteristics of the known material and the unknown material, the relationship being based on measurement data of both the substance and the known material.

30. The method of claim 18, wherein The object comprises three or more different material components, four or more different material components, or five or more different material components.

31. The method of claim 18, wherein The number of energy levels (n) is less than, equal to, or greater than the number of materials to be decomposed and / or identified.

32. The method of claim 18, wherein: Reducing the effects of X-ray scatter from detected X-rays includes using interpolation to obtain a high-resolution primary image or using a beam selector or collimator to measure a low-resolution primary image.

33. The method of claim 18, further comprising using dual energy decomposition to generate the at least one image and / or identify the at least one substance.

34. The method of claim 18, further comprising generating the at least one image and / or identifying the at least one substance using decomposition by a linearization method.

35. The method of claim 18, wherein The object is irradiated with broad spectrum X-rays, and the received X-rays are measured using energy sensitive, photon counting and / or time sensitive methods.

36. The method of claim 18, further comprising performing a calibration method and / or calibration step for scatter removal.

37. The method according to claim 36, wherein Calibration is performed using each material composition and / or a composite of each material composition having various thicknesses.

38. The method of claim 18, further comprising establishing a database for each material component and each combination of the material component with two or more components at each energy level by measuring values ​​with no or low scattering interference, and using the database to identify the at least one substance.

39. The method according to claim 38, wherein The physical properties of the substance are established by using measurements of the X-ray measurement device, or are characterized by using simulations or synthetic data.

40. The method of claim 39, wherein The relationship between the measurement of the known material and the substance to be measured involves the density, atomic z and / or thickness of the respective material and substance.

41. The method of claim 18, wherein The at least one image is generated by referencing a database of stored CT, MRI, SPECT, PET, spectral, photoacoustic and photoenergetic, chemical and / or electrochemical measurements.

42. The method of claim 28, wherein The database includes one or more of the following: physical, chemical, electrical, physiological and / or dynamic properties, at least one phase of the dynamic process and / or movement of the known material in static, temporal, dynamic and / or spatial form.

43. The method according to claim 42, wherein The properties include one or more of characteristic regions, atomic z numbers, densities, unstructured molecules, structured molecules, synthetic molecules, synthetic materials, air, microstructures, structures in the region of interest having at least one marker in physical, chemical, spatial, temporal and physiological form.

44. The method of claim 42, wherein: The known material is one or more of: a tissue region, a contrast-marked plaster cast, a material and cavity form, one or more regions of an organ, a molecular complex, a contrast-marked molecular complex, a contrast-marked organelle, an ion mixture, an ablated region of tissue, a tumor, a diseased region of tissue, a chemical compound, a semiconductor component, a metallic portion, an inorganic or organic substance and / or a mixture of inorganic and organic substances.

45. The method according to any one of claims 18 to 27, wherein After the X-rays have passed through the object, detecting the X-rays by the X-ray measurement device includes determining interference patterns, phase contrast imaging, coherence and partial coherence X-ray imaging measurements.

46. ​​The method of claim 19, wherein The at least one substance is a unique and rare composition material, and the method further includes a second order approximation method and a second decomposition method.

47. The method of claim 46, wherein The unique and rare component material is a tissue, modified tissue, calcification, microcalcification and / or contrast agent marked substance, surgical tool, biopsy tool, therapeutic tool and / or implant.

48. The method of claim 47, wherein The treatment tool is a catheter, an RF ablation probe, a guide wire, an optical biopsy needle and / or a radiotherapy probe.

49. The method of claim 47, wherein The contrast agent-labeled substance is an iodinated agent, barium sulfate and / or a CT or X-ray sensitive contrast agent, nanoparticles, particles with an atomic z different from the rest of the region of interest, and / or particle derivatives combined with molecular markers.

50. The method of claim 19, further comprising using a k-edge method to identify the at least one substance.

51. The method of any one of claims 19 to 27, further comprising using an artificial intelligence method, a deep machine learning method, an artificial neural network method, a convolutional neural network method, and a deep neural network method to identify the at least one substance.

52. The method according to any one of claims 19 to 27, wherein The at least two-dimensional detector is a flat panel detector.

53. The method of claim 19, wherein: The at least two-dimensional detector is a photon counting detector, a photomultiplier tube (PMT), a light counting diode, a diode array, an energy sensitive detector and / or an energy sensitive spectrometer.

54. The method of claim 18, wherein The object is irradiated with X-rays having one or more energy peaks in at least one pulse, and the received X-rays are measured using energy-sensitive, photon-counting, and / or time-sensitive methods.

55. The method of claim 18, wherein At least one image is generated through user or digital program settings used in artificial intelligence or neural networks.

56. The method of claim 18, comprising reducing X-ray dose via a closed-loop feedback system.

57. A system for imaging an object, the system comprising: an X-ray source configured to irradiate the object with X-rays having at least two different energy levels; an X-ray measuring device having at least one two-dimensional detector and configured to enable an object to be positioned between the X-ray source and the X-ray measuring device, the X-ray measuring device being further configured to detect the X-rays after the X-rays have passed through the object; wherein the radiographic X-ray measurement device is configured to reduce the effects of X-ray scattering from the detected X-rays to generate at least one image, and wherein the X-ray measurement device is configured to perform a pre-measurement calibration using each material component and a composite of each material component having various thicknesses.

58. The system of claim 57, wherein: The X-ray measurement device is further configured to identify at least one substance in the at least one image.

59. The system of claim 58, wherein: The X-ray measurement device is further configured to identify a plurality of different substances in the image.

60. The system of claim 59, wherein: The X-ray measurement device is further configured to identify substances in the at least one image that are indicative of an abnormal condition, a pathological state, and / or at least one change in a pathological state.

61. The system of claim 59, wherein: The substances are microcalcifications.

62. The system of claim 59, wherein: The substance is a cancer, a tumor, a contrast agent indicative of a tumor, a nanoparticle, an iodinated blood vessel, and / or a contrast agent indicative of a tissue.

63. The system of claim 62, wherein the tumor is indicative of lung cancer.

64. The system of claim 59, wherein: The X-ray measurement device is further configured to identify the substance using a database of established algorithmic relationships between known materials, the substance to be detected, and / or known materials similar to the substance.

65. The system of claim 58, wherein: The object comprises three or more different material components, four or more different material components, or five or more different material components.

66. The system of claim 58, wherein: The number of energy levels (n) is equal to or greater than the number of materials to be resolved and / or identified.

67. The system of claim 58, wherein: The X-ray measurement device is further configured to use interpolation to reduce the influence of X-ray scatter from the detected X-rays.

68. The system of claim 58, wherein: The X-ray measurement device further uses dual energy decomposition and / or linear method decomposition to generate the at least one image and / or identify the at least one substance.

69. The system of claim 58, wherein: The X-ray source irradiates the object with broad spectrum X-rays, and the X-ray measurement device is further configured to receive the X-rays for measurement using an energy sensitive or photon counting method.

70. The system of claim 58, wherein: The X-ray measurement device is further configured to perform the pre-measurement calibration for scatter removal.

71. The system of claim 70, wherein: Calibration was performed using each material composition and composites of each material composition having various thicknesses.

72. The system of claim 58, further providing the X-ray measurement device with a database for each material component and each combination of the material component with two or more components at each energy level by derived measurements without scatter interference, and using the database to identify the at least one substance.

73. The system of claim 72, wherein: The database provides a reference for measurements created using the X-ray measurement device or using simulated or synthetic data.

74. The system of claim 73, wherein: The identification and determination of properties and / or characteristics of the substance is established by measurements using the X-ray measuring device and / or using simulated, synthetic and / or predetermined data.

75. The system of claim 74, wherein: The relationship between the measurement of a known material and the substance to be measured involves the density, atomic z and / or thickness of the respective material and substance, wherein the known material and the substance have similar X-ray attenuation properties.

76. The system of claim 58, wherein: The at least one image is generated by referencing a database of stored CT, MRI, SPECT, PET, spectroscopy, photoacoustic and other energy, chemical and / or electrochemical measurements.

77. The system of claim 76, wherein: The database comprises one or more of the following: physical, chemical, electrical, physiological and / or dynamic properties, dynamic processes of known materials in static, temporal, dynamic and / or spatial form and / or at least one phase of motion.

78. The system of claim 77, wherein: The properties include one or more of characteristic regions, atomic z numbers, densities, unstructured molecules, structured molecules, synthetic molecules, synthetic materials, air, microstructures, structures in the region of interest having at least one marker in physical, chemical, spatial, temporal and physiological form.

79. The system of claim 78, wherein: The known material is one or more of: a tissue region, a contrast-marked plaster cast, a material and cavity form, one or more regions of an organ, a molecular complex, a contrast-marked molecular complex, a contrast-marked organelle, an ion mixture, an ablated region of tissue, a tumor, a diseased region of tissue, a chemical compound, a semiconductor component, a metallic portion, an inorganic or organic substance and / or a mixture of inorganic and organic substances.

80. The system of claim 58, wherein The X-ray measurement device is configured to determine measurements of interferograms, phase contrast imaging, coherence and partial coherence X-ray imaging.

81. The system of claim 59, wherein: The substance is a unique and rare composition material and alternative methods including a second order approximation method are employed.

82. The system of claim 81, wherein The unique and rare component material is a tissue, a tissue-modified and / or contrast-labeled substance, a surgical tool, a biopsy tool, or an implant.

83. The system of claim 82, wherein: The contrast agent is an iodinated agent, barium sulfate and / or CT or X-ray contrast material or nanoparticles and derivatives thereof.

84. The system of claim 59, wherein: The X-ray measurement device is further configured to identify the substance using a k-edge method.

85. The system of claim 59, wherein: The X-ray measurement device and / or the linear or dual energy decomposition method is also configured to identify the substance using artificial intelligence methods, deep machine learning methods, artificial neural network methods, convolutional neural network methods and deep neural network methods.

86. The system of claim 59, wherein: The X-ray measurement device and material decomposition method are also configured to identify and distinguish substances based on facts derived from one or more criteria or algorithms among artificial intelligence methods, deep machine learning methods, artificial neural network methods, convolutional neural network methods and deep neural network method algorithms.

87. The system of claim 59, wherein: The at least one two-dimensional detector is a flat panel detector.

88. The system of claim 59, wherein: The at least one two-dimensional detector is a photon counting detector, a PMT, a light counting diode, a diode array, an energy sensitive detector and / or an energy sensitive spectrometer.

89. The system of claim 77, wherein: A subset of the database includes structured or unstructured data constructed based on the results of artificial intelligence methods, deep machine learning, artificial neural networks, convolutional neural networks and / or deep neural network methods.

90. The system of claim 89, wherein: The database is referenced based on one or more criteria set by a user or a digital program.

91. The system of claim 90, wherein: The criteria include one or more of algorithms, artificial intelligence, deep machine learning, artificial neural networks, convolutional neural networks, and deep neural networks.

92. The system of claim 57, wherein the system is portable.

93. The system of claim 57, wherein the X-ray source is battery powered.

Images