X-ray imaging with energy sensitivity

Abstract

An X-ray imaging system is provided that includes an X-ray source and an energy-resolving X-ray detector. The X-ray detector includes a scintillator and a photodetector configured to detect X-rays as a function of position and energy simultaneously. The X-ray imaging system is configured to generate energy and / or phase-resolved X-ray images. Some embodiments include a scintillator that converts high energy photons into UV light, the scintillator coupled with an array of photodiodes, the output of which is proportional to the amount of light generated. In some embodiments, the scintillator includes GAGG: Ce (cerium doped (gadolinium aluminum gallium) garnet). In some embodiments, the X-ray detector is configured to use quantum noise or quantum scattering. The resulting phase-contrast and / or energy-resolved images may produce clearer and clearer images relative to conventional CT.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is a continuation-in-part of U.S. non-provisional patent application serial number 17,579,481 filed on January 19, 2022, which in turn is a continuation-in-part of U.S. non-provisional patent application serial number 16 / 591,581 filed on October 2, 2019, which in turn is a continuation-in-part of U.S. non-provisional patent application serial number 62 / 739,954 filed on October 2, 2018. The disclosures of these patent applications are hereby incorporated herein by reference. Background Art Technical field:

[0003] The present application belongs to the field of X-ray imaging.

[0004] Related technologies:

[0005] The major health concern with radiography, including computed tomography, is the cumulative total dose of ionizing radiation. Reliable data on radiation exposure and associated health effects associated with medical imaging are lacking, but radiation is known to cause cancer from very reliable data from large populations exposed to almost instantaneous 10 rad (0.1 sievert) exposures (survivors of Hiroshima and Nagasaki, Japan), but it is not known how this scales down to cumulative doses received over long periods of time, as is typical in medical imaging. Mathematically, the functional relationship between radiation carcinogenesis and actual dose remains unknown.

[0006] Given this situation, a prudent approach is to obtain the required imaging for medical diagnostic purposes with the least amount of radiation exposure. Limiting the total dose is the standard approach in this regard, but runs into the problem of quantum mottle, which consists of granularity in the image resulting from statistical fluctuations in adjacent pixels. These fluctuations are related to (N) -1. / 2 proportional to N, and if N is not large enough to drive such fluctuations to ~0, the human eye will interpret these fluctuations as changes in grayscale, making the image look grainy and very difficult to interpret.

[0007] Traditionally, before the advent of computed tomography (CT), chest imaging was strictly radiographic using a wide-angle X-ray beam and a single exposure with no variation in the angulation of the beam. In contrast, CT uses many narrow beam exposures arranged over 360 degrees. The typical dose in non-CT radiographic techniques is about 0.03 mSv or 30 mrad. Digital non-emulsion technology has reduced this to about 0.015 mSv or 15 mrad. Therefore, even low-dose chest CT requires the patient to receive 10 times the X-ray dose of radiography.

[0008] The typical dose in lung cancer screening CT is about 1.5 rads per study visit. While one can get very good images of the lungs themselves at much lower doses (e.g., at the 0.60 rad level), such low doses make the chest wall and surrounding tissues so grainy that they become difficult to interpret. Because legal standards make radiologists responsible for everything on each image, a low-dose technique that might be adequate to identify lung cancer can mask cancer in nearby tissue (e.g., in the fornix of the liver).

[0009] A significant drawback of conventional imaging techniques is that virtually all X-ray and gamma-ray medical imaging is done using energy-blind detectors. Such systems combine (i.e., integrate at each pixel) photons of all energies into the final image without making any useful distinction between the photons' energies. Summary of the invention

[0010] Various embodiments include systems and methods for generating energy-resolved X-rays. These systems can provide several significant advantages. First, detection of energy-resolved X-rays enables photon phase detection and thereby allows for phase-resolved images. Second, detection of energy-resolved X-rays enables improved statistical analysis of received X-rays, which enables energy-dependent smoothing functions and quantum speckle suppression. Third, detection of energy-resolved X-rays enables multi-energy (e.g., dual-energy or triple-energy) techniques that are desirable for tissue identification without the need for multiple, individually energy X-ray sources and individual detectors.

[0011] Various embodiments of the invention include an imaging system comprising: an X-ray source configured to generate X-rays having a range of energies; an X-ray detector comprising an array of pixels and configured to quantify the energy of the detected X-rays, each of the pixels comprising: a scintillation crystal configured to generate ultraviolet light in response to the X-rays, the amount of the ultraviolet light being dependent on the energy of the photons of the X-rays; an ultraviolet light sensitive photon detector configured to generate an electrical signal proportional to the amount of the ultraviolet light; and electronics configured to quantify the electrical signal, the quantized electrical signal representing the energy of the X-rays; and image generation logic configured to generate at least a first image based on the position of each of the pixels within the array and the energy of the X-rays detected at each of the pixels. Optionally, the scintillation crystal comprises gadolinium aluminum gallium garnet (GAGG) optionally doped with cerium. It is contemplated that other GAGG scintillators and / or versions may be used in alternative embodiments. Optionally, the imaging system further includes a contrast agent to be placed in the object to be imaged and configured to absorb photons having a selected set of X-ray energies, wherein the image generation logic is configured to generate an image that emphasizes or deemphasizes (or otherwise manipulates) photons having the selected set of X-ray frequencies.

[0012] Various embodiments of the present invention include a method of producing an X-ray image, the method comprising: generating X-rays using an X-ray source; passing the X-rays through an object to be imaged; detecting the X-rays using an energy-resolving detector, the energy-resolving detector having a two-dimensional array of pixels, each of the pixels being configured to determine the energy of each received X-ray photon, such that X-ray data representing both the two-dimensional position and energy of each X-ray is generated, wherein the energy of the X-rays is classified into at least three energy bins; and generating at least a first image based on the X-ray data. Optionally, the method further comprises applying a smoothing function to the X-ray data, the smoothing function being configured to reduce quantum speckles in the first image and depending on the energy of the received X-ray photons; and / or wherein the first image is an energy-resolved image in which X-ray photons of a first energy are highlighted or faded relative to photons of a second (or more) energy. The first image may optionally depend on and / or be modified by the measured photon energy information within a single exposure. Optionally, the method further comprises selecting photons of energy absorbed by a predetermined contrast agent or alternatively not absorbed by the predetermined contrast agent, and generating a second image based on highlighting or de-emphasizing or otherwise manipulating the selected photons. For example, an image may be generated that highlights locations where an iodine contrast agent absorbs photons within a specific energy range. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The use of an energy-resolved X-ray imaging system according to various embodiments of the present invention is demonstrated, the energy-resolved X-ray imaging system being configured for generating energy-resolved X-ray images of a patient.

[0014] Figure 2 The energy spectra of X-ray tubes according to different embodiments of the present invention are shown.

[0015] Figure 3 Cross-sectional views of pixels of an energy resolving X-ray detector according to different embodiments of the present invention are shown.

[0016] Figure 4A and Figure 4B Isometric and top views of an energy resolving X-ray detector including a pixel array according to different embodiments of the present invention are shown.

[0017] FIG. 5A to FIG. 5C Manipulation of phase resolved X-ray and / or energy resolved data to image specific tissues according to various embodiments of the present invention is demonstrated.

[0018] Fig. 6A and Figure 6B Selection of soft tissue data in energy-resolved X-rays according to different embodiments of the present invention is demonstrated.

[0019] Fig. 7A and Figure 7B The differences between a standard "dual energy" CT system and an energy resolved CT system according to various embodiments of the present invention are demonstrated.

[0020] Fig. 8A and Figure 8B The use of an energy resolving detector to select calcified tissue in breast X-rays according to various embodiments of the present invention is demonstrated.

[0021] Fig. 9 Methods of generating images using an energy resolving X-ray detector according to various embodiments of the present invention are presented.

[0022] The drawings disclosed herein are generally not drawn to scale. DETAILED DESCRIPTION

[0023] Figure 1The use of an energy-resolved X-ray imaging system 100 according to various embodiments of the present invention is demonstrated, which is configured to generate an energy-resolved X-ray image of a patient. The energy-resolved X-ray image is based on X-ray data that includes both position (x, y) and energy information. The energy information is optionally implemented in an array of energy bins, each energy being associated with a predetermined energy range. Thus, a specific detected X-ray photon can be characterized by at least three values: X, Y, and E, where E is represented by a specific energy bin, and X and Y refer to the two-dimensional position of each pixel in the detector pixel array. Different embodiments may include 2-64 or more energy bins, or any range between 2 and 64 bins. The number of bins is determined in part by the detection electronics. Such electronics may support 128, 256, 512, or 1024 or more bins.

[0024] like Figure 1 As shown in FIG. 1 , the X-ray imaging system 100 includes an X-ray source 110 configured to generate X-rays having a certain energy range. The X-ray source 110 may include a variety of known X-ray sources, including, for example, an X-ray tube, a synchrotron radiation source, a radionuclide, and / or any other X-ray source known in the art. In some embodiments, the X-ray source 110 is configured to generate X-rays having a certain energy range. Figure 2 2. Such an X-ray spectrum 210 may include characteristic radiation associated with a particular element, such as iodine.

[0025] The imaging system 100 optionally further includes phase selectors 120A and 120B. Phase selectors 120A and 120B are configured to select photons with a specific phase shift or photons with a specific phase difference. For example, phase selector 120A can be adjusted to pass photons of a specific phase, while phase selector 120B is adjusted to pass photons of the same phase or slightly different phases (e.g., a phase difference of 5 degrees). In this example, photons passing through phase selector 120A and phase selector 120B must undergo a phase shift of ~5 degrees between phase selectors 120. Specifically, they must be phase shifted by interaction with object 130. In various embodiments, phase selector 120 is configured to distinguish between photons having a phase difference of a few degrees (e.g., at least 2, 3, 5, or 10 degrees) or any range therebetween.

[0026] Phase selectors 120A and 120B may include gratings or other wavelength-dependent optical devices. Typically, at least one of phase selectors 120A and 120B includes a phase selection adjustment mechanism, such as a mechanical dial, a motor (i.e., a stepper motor), a grating rotator, and / or the like, which is configured to change the relative phase passed by phase selectors 120A and 120B, i.e., pass photons with a selected phase shift. By adjusting the mechanism, the range of the phase difference (phase shift) between the relative phases of phase selectors 120A and 120B can be scanned.

[0027] The X-rays generated by the X-ray source 110 pass through an area where a subject 130 may be placed. The subject 130 may include a patient or other object to be X-rayed. The space may include a structure configured to support the subject 130, such as a platform.

[0028] The object 130 can absorb and / or change the phase of the X-ray photons generated by the X-ray source 110. For example, photon absorption can vary with both tissue / material density and photon energy. In this way, distinguishing the absorption of photons of different energies can allow tissues / materials of different densities to be more clearly distinguished. The phase of the X-ray photons can be changed by the interaction of the X-rays at the boundaries between materials of different densities and / or refractive indices (at the X-ray wavelength). Therefore, adjusting the phase selectors 120A and 120B to detect different changes in phase caused by the object 130 can be used to detect boundaries between different tissues and / or material types. In a specific example, phase-resolved X-ray data can be used to more clearly (relative to non-phase-resolved X-ray data) image vascular boundaries or boundaries of a tumor.

[0029] The imaging system 100 further includes an energy-resolving detector 150. The detector 150 includes an array of pixels and is configured to quantify the energy of the detected X-rays. In different embodiments, the pixel array can be linear or two-dimensional. For example, the detector 150 can include an array of at least 200×200, 256×256, 512×512, or 1024×1024 pixels. In different embodiments, the detector 150 includes at least 40k, 64k, 262k, 1M, 4M, or 8M pixels, or any range therebetween. The pixel array can be flat or curved. For example, in some embodiments, the pixel array is arranged into a set of flat hexagons, which are in turn arranged to form a curved (e.g., parabolic) shape.

[0030] The imaging system 100 optionally further includes a filter 140 configured to select an energy range of the X-rays received by the object 130. The filter 140 may include a bandpass filter or a notch filter. In some embodiments, the filter 140 is configured to pass X-rays with energies expected to come from a certain element, such as iodine. For example, the filter 140 may be configured to pass X-rays with energies expected to come from an element, such as iodine. Figure 2 The iodine peak in the X-ray spectrum shown in FIG.

[0031] Figure 3 A cross-sectional view of a pixel 310 that may be included in an energy-resolving X-ray detector 150 according to various embodiments of the present invention is shown. In a typical embodiment, each pixel 310 includes at least a scintillation crystal 320, an optical diode 340, and an electronic device. The scintillation crystal 320 is configured to generate ultraviolet light in response to an X-ray. Specifically, when the X-ray irradiates the scintillation crystal 320, ultraviolet light and optional visible light are generated. The amount of ultraviolet light depends on the energy of the photon of the detected X-ray. Photons with higher energy produce more light relative to photons with lower energy. As used herein, the term "energy resolution" is intended to indicate that the detector is configured to distinguish between X-rays of different energies, optionally assigning each X-ray photon to an energy bin in an array of alternative energy "bins". For example, each energy bin may represent an energy range of 0.1keV, 0.5keV, 1keV, or 2keV. In some embodiments, the number of energy bins is selected to include at least 8 or 10 bins to cover the fwhm (full width at half maximum) of the X-ray output spectrum of the X-ray source 110. The alternative number of energy bins is discussed elsewhere herein. In some embodiments, the scintillator crystals 320 include gadolinium aluminum gallium garnet (GGAG), optionally doped with cerium. Fast, bright, and / or balanced GGAG can be used in different embodiments. In alternative embodiments, the scintillator crystals 320 are supplemented or replaced by cadmium zinc telluride (CZT) and / or other scintillators known in the art. For example, some embodiments include CZT disposed between the optical diode 340 and the GGAG layer. The GGAG is optionally configured to provide a moisture barrier.

[0032] Optical diode 340 is an ultraviolet light sensitive photon detector (e.g., a photodiode) configured to detect ultraviolet light generated within scintillation crystal 320. In response to the ultraviolet light, optical diode 340 is configured to generate an electrical signal proportional to the amount of ultraviolet light. As such, the electrical signal is also proportional, optionally linearly proportional, to the energy of the detected X-ray photons. In alternative embodiments, optical diode 340 may be replaced by an optical transistor or other optoelectronic device.

[0033] The electronics of the detector 150 are configured to quantize (e.g., digitize) the electrical signal generated by the optical diode 340. The quantized electrical signal represents the energy of the detected X-ray. The electronics optionally include a front-end electronics 350 and a back-end electronics 360. The front-end electronics 350 typically include circuits such as amplifiers. The back-end electronics 360 may include circuits such as a pulse height analyzer or a pulse integrator, and a memory configured to temporarily store digital results. Typically, each pixel 310 includes a dedicated optical diode 340 and a front-end electronics 350, and optionally a dedicated back-end electronics 360. In some embodiments, each pixel 310 includes a dedicated scintillation crystal 320, but in alternative embodiments, a single scintillation crystal 320 is used to convert X-rays into ultraviolet light so that it can be detected by multiple optical diodes 340.

[0034] The pixel 310 optionally further includes an optical coupler 330 located between the scintillator crystal 320 and the optical diode 340. The optical diode 340 and / or associated front-end electronics 350 are optionally produced in a pre-configured array and then incorporated into the pixel 310. For example, multiple optical diodes 340 can be produced in an array on a single silicon substrate before being incorporated into the pixel 310 array.

[0035] Reference again Figure 1 , the imaging system 100 optionally further includes image generation logic 160. The image generation logic 160 is configured to generate an image based on the position of each pixel 310 within the array and also based on the energy of the X-rays detected by each pixel 310. Typically, the image is based on the detection of a large number of X-ray photons, each X-ray photon being associated with an X / Y position (e.g., a position in two dimensions) and an energy E. The image generation logic 160 includes electronic circuitry, hardware, firmware, and / or software stored on a non-transitory computer-readable medium. For example, in some embodiments, the image generation logic 160 includes one or more computing devices that include image generation software and an input / output configured to receive energy-resolved X-ray data generated by the detector 150.

[0036] In some embodiments, image generation logic 160 is configured to generate at least a second image based on photons having different phase shifts. Data for generating such an image may be obtained by detecting X-rays at a first setting of phase selectors 120A and 120B and then detecting X-rays at a second setting of phase selectors 120A and 120B. The difference between the data generated at these two settings may be used, for example, to generate an image that highlights the edges of structures within object 130.

[0037] In some embodiments, the image generation logic 160 is configured to generate a set of images based on photons of different energies, or to generate one or more images in which the X-ray energy is represented by false colors. Optionally, the image generation logic 160 includes a user interface in which a user can select one, two or more energy ranges and manipulate the image data in each energy range independently. For example, a user can select two energy ranges and highlight or fade the photons of different energies within these ranges. This can be done by reversing, subtracting and / or adding data in different energy ranges. In a specific example, photons in a higher energy range can be absorbed by denser materials (e.g., tumors, bones, or calcified lymph nodes) relative to lower energy photons absorbed by less dense materials (e.g., lung tissue, muscle, adipose tissue). By subtracting X-ray data of higher energy X-rays from the image and / or adding X-ray data of lower energy X-rays, the details of less dense materials can be made more apparent in the resulting image. This provides significant clinical benefits.

[0038] The user interface may include features such as: 1) rapidly alternating between images to produce differential "flickering" to highlight features resulting from absorption of X-rays of different energies; and / or 2) allowing a user to select an energy range to be highlighted or deemphasized using a virtual slider bar or dial, thereby dynamically manipulating the X-ray data to selectively inspect and / or search for specific structures. In some embodiments, the image generation logic 160 includes a neural network and / or machine learning system configured to automatically process image data according to energy (i.e., according to energy quantized in (at least 3, 4, 8, 10, etc.) energy bins as discussed elsewhere herein). The output of this processing may include a composite image in which different structures within the object 130 are shown in their most distinctive appearance. For example, if an image includes calcified lymph nodes, bones, tumors, and fatty breast masses, each of these structures may be most clearly seen at different specific energy levels. The composite image may include data points representing each of these structures as detected at the optimal X-ray energy, thereby producing an optimized image for efficient analysis by the user. Computer-aided detection (CAD) may be used to generate a composite image to detect specific objects, and then select energies to highlight or deemphasize the energies that most clearly see and / or distinguish the detected objects.

[0039] In some embodiments, the image generation logic 160 is configured to apply an energy-dependent smoothing function to the X-ray image. The smoothing function is optionally configured to reduce quantum speckles in the image. The energy information produces an improved smoothing function because the effect of quantum speckles is energy-dependent. When the energy of the X-ray photons detected by the detector 150 is known, the improved smoothing function is applied. For example, when the energy of the X-ray photons is measured, the actual energy spectrum of the received photons can be generated and the energy spectrum can be used to provide a best fit through the data, optionally imaging pixel by pixel, thereby reducing statistical fluctuations. The improved smoothing function means that the object 130 can be exposed to a lower total dose of X-rays to obtain a balanced image, or an improved image can be obtained at the same dose. For example, when the expected energy distribution is known, it is easy to see the changes in the energy spectrum, and therefore estimate whether statistically sufficient samples have been obtained. In some embodiments, the image generation logic 160 is configured to terminate the generation of X-rays from the X-ray source 110 based on receiving statistically sufficient data on the energy bin.

[0040] In various embodiments, quantum speckle suppression significantly improves the quality of standard chest radiographs and all radiography by digitizing each photon (X-ray or gamma ray) one at a time using a detector that can count and record the position and energy of each photon. This is very useful not only for standard medical imaging, but also for advanced imaging using phase contrast techniques. Phase contrast techniques can use radiographic techniques to provide clearer images than those obtainable in CT, and the energy of each photon helps determine the photon phase or phase shift. The technique is based on noting which photons interact with the edges of an object and are therefore delayed (phase shifted) on their way to the detector, and since photons propagate as waves, the phase of such delayed photons is different from the phase of other photons. Phase contrast imaging can avoid the need for CT in many cases, which has many advantages in addition to the promise of lower calculated radiation doses compared to CT.

[0041] Figure 4A and Figure 4BAn isometric view and a top view of an energy-resolving X-ray detector including a pixel array 410 according to various embodiments of the present invention are shown. The scintillator crystals 320 are optionally grown on an X-ray transparent substrate 420. The scintillator crystals 320 of each pixel 310 may be combined in a single crystal. Alternatively, a barrier 430 may be disposed between the scintillator crystals 320 of each pixel 310. The barrier 430 is typically grown with the scintillator crystals 320 and may include, for example, a metal, or a crystal region that has been grown slightly differently to produce a change in refractive index at ultraviolet and / or X-ray frequencies, for example, annealed or doped differently. The barrier 430 is configured to prevent ultraviolet and / or X-rays from propagating between the scintillator crystals 320. Since the number of photons generated by the scintillator is proportional to the absorbed X-ray energy, the same energy resolution can be achieved over a wide range of scintillator thicknesses.

[0042] FIG. 5A to FIG. 5C Manipulation of phase resolved and / or energy resolved X-ray data to image specific tissues according to various embodiments of the present invention is demonstrated. Figure 5A is a conceptual diagram of tissue 510 including vein 520. The edges of the vein can be distinguished by X-ray phase shift, and / or the vein can be distinguished by contrast agent. Figure 5B In , the X-ray data is manipulated to highlight tissue 510 and de-emphasize veins 520. Figure 5C , the X-ray data is manipulated to highlight veins 520 and de-emphasize tissue 510 .

[0043] Fig. 6A and Figure 6B The selection of soft tissue data in energy-resolved X-ray according to different embodiments of the present invention is shown. Fig. 6A In the image, a dense bone structure 610 is distinguished. This structure can be seen more clearly in data collected from higher energy X-rays. Figure 6B , soft tissue structures 620 are distinguished. Such structures can be seen more clearly in data collected from relatively low energy X-rays. As noted herein, a user can choose to highlight or de-emphasize such different structures in an energy-resolved image (an image manipulated based on detected X-ray energy).

[0044] Fig. 7A and Figure 7BThe difference between a standard "dual energy" CT (computed tomography) system 710 (such as can be found in the prior art) and an energy-resolved CT system 760 according to different embodiments of the present invention is shown. In the dual energy CT system 710, two X-ray sources are used: a high energy X-ray source 720 and a low energy X-ray source 730. Each of these sources is pre-configured to provide X-rays in a different energy range. The two X-ray sources are optionally mounted on a gantry 740 configured to move the X-ray sources in order to generate CT data. The X-rays from the two sources are directed to an object 130, which is optionally disposed on a platform 755. The X-rays from each source are detected by a separate (energy blind) X-ray detector 735.

[0045] This dual-energy CT system 710 has some advantages. For example, dual-energy CT identifies calcium and ossification better and more accurately than standard CT. However, the dual-energy CT system 710 has many disadvantages. Such a system requires two detectors, two X-ray sources, and a much more expensive CT machine (partly due to the more powerful gantry 740), which uses two X-ray sources and two detectors separated by -90° from each other to take two separate but simultaneous images of the object. The two X-ray sources are set to different tube energies, resulting in very poor energy selection between the two tube / detector systems, and these separate images are "subtracted" to better identify calcium and ossified nodules. The object is exposed to a double dose of X-ray radiation, and the energy of each X-ray source is fixed when collecting data. There are only two energy data sets ("bins"), each of which is associated with a separate data set. This technology increases the dose to the patient by about two times, and the energy selection / differentiation is poor because the detector is actually energy blind.

[0046] Compared with the dual-energy CT system 710, the energy-resolved CT system 760 (see Figure 7B ) requires only a single X-ray source 770 and a single energy sensitive / selective detector, such as X-ray detector 150. X-ray source 770 is optionally an embodiment of X-ray source 110. Indeed, in some embodiments, existing CT systems can be retrofitted with X-ray detector 150 to convert into energy resolving systems. These systems can then benefit from multi-energy data and energy-based data manipulation as described herein, with a single image and lower dose than a standard dual energy CT system 710. By simply splitting the energy spectrum of photons into high and low (or 3 or more) energy bins, "dual energy" or "multi-energy" can be obtained.

[0047] Fig. 8A and Figure 8B The use of an energy-resolving detector to select calcified tissue in breast X-rays according to various embodiments of the present invention is demonstrated. Fig. 8AIn FIG. 8 , a breast image 810 includes X-ray signals generated by structures within the breast and soft tissue 830 of the breast. Figure 8B 8, breast image 840 includes X-ray (signal) data that downplays energy that will be absorbed by soft tissue and X-ray data that highlights energy that will be absorbed by structures within the breast (eg, calcified tissue 850 and / or tumor 860).

[0048] It is known that dual energy CT is useful in tissue identification. This is important in the case of cancer treatment. For example, one of the main ways to monitor the progress of cancer and cancer treatment is to monitor metastasis to lymph nodes. Effective treatment destroys metastatic cancer in lymph nodes. In the case of complete eradication of cancer from lymph nodes, lymph nodes usually do not completely return to their original state, but calcify. The lymph node itself dies. This is called a "granuloma". The particles (i.e. calcium) containing nodules. The distinction between cancerous lymph nodes and calcified lymph nodes can be completed using detector 150 to generate energy and / or phase resolved images as discussed herein.

[0049] Fig. 9 Methods of generating images using an energy resolving X-ray detector according to various embodiments of the present invention are presented. Fig. 9 The steps shown in may optionally be performed in alternative orders as well as simultaneously, and may be performed using the X-ray imaging system 100 discussed elsewhere herein.

[0050] In the generate X-ray step 910, X-rays are generated using an X-ray source such as the X-ray source 110. The generated X-rays may have Figure 2 In some embodiments, once a statistically relevant number of X-rays are detected in a plurality of energy bins, the generate X-ray step 910 is automatically terminated.

[0051] In passing X-rays step 920, the generated X-rays pass through an object to be examined, such as object 130. The object can be any object for which X-ray imaging is currently performed, including, for example, a person, an animal, an artifact, a historical artifact, an assembled or manufactured material, a geological sample, etc. The object can contain a contrast agent, such as barium or iodine, which can be used as a contrast agent in medical applications. Within the object, the X-rays may be attenuated, absorbed, changed phase, scattered, etc.

[0052] In a detect x-ray step 930, an energy-resolving detector (such as detector 150) is used to detect x-rays that have passed through the object. The energy-resolving detector is configured to determine the energy of each received x-ray photon. The energy-resolving detector optionally includes a two-dimensional array of pixels, such as an array of pixels 310, so that x-ray data is generated that simultaneously represents the two-dimensional position and energy of each x-ray. In a typical embodiment, the energy-resolving detector is configured to classify the x-rays into at least 2, 3, 4, 8, 10, 16, 32, or 64 energy bins, or any range therebetween.

[0053] In an optional detect phase step 940, a phase shift of each received X-ray photon is detected. Phase shifts are optionally detected by setting the phase selector 120 to different configurations so as to pass photons having a specific phase difference (e.g., approximately 0 degrees, 2 degrees, 4 degrees, 5 degrees, or 6 degrees). X-ray data may be obtained at one, two, or more settings of the phase selector 120 (each setting configured to cause X-rays to reach the detector at a different phase shift). As noted elsewhere herein, phase shifts may be caused by the interaction of X-rays at edges within the object. Detect phase step 940 is optionally performed simultaneously with detect X-ray step 930. In some embodiments, edge detection is used to distinguish between structures having cancerous shapes and non-cancerous shapes.

[0054] In a generate image step 950, one or more images are generated based on the X-ray data generated in the detect X-ray step 930. The generated images may include, for example, images of X-rays based on a specific X-ray energy or energy range, images of X-rays based on a specific phase shift or phase shift range, and / or images that highlight and / or deemphasize X-ray data based on energy and / or phase. For example, an "energy resolved" image may be generated by subtracting data associated with one energy from a data set or from data associated with a second energy. An image may be generated by subtracting data associated with one phase shift from data associated with a second phase shift. In a more specific example, data with energy that will be absorbed by the contrast agent is highlighted relative to data with energy that will not be absorbed by the contrast agent. In another example, data showing a phase shift is highlighted relative to data with no phase shift or with minimal phase shift.

[0055] In some embodiments, the generate image step 950 includes automatically generating an image using computer-aided detection (CAD). For example, a machine learning system can be trained to manipulate data of different X-ray energies in order to automatically highlight specific structures within an object. Further, the generate image step 950 can include generating a composite image in which different areas of the image are manipulated in different ways. For example, a first portion of the image can be manipulated to highlight a first structure, while a second portion of the image can be manipulated to highlight a second structure, which structures optionally have different features associated with X-ray interactions. In some embodiments, the generate image step 950 includes adding false color to the image based on the energy and / or phase shift of the detected X-rays.

[0056] In some embodiments, the generate image step 950 includes user manipulation of controls such as a virtual dial or slider bar to dynamically select energies and / or phase shifts for manipulating the X-ray data used to generate the image. For example, the user can use the controls to select X-ray energies to be highlighted and / or de-emphasized, and then observe the resulting image in real time. This selection can be from 1, 2, 3, 4, 8, 10, 16, 32, or 64 energy bins, or more, or any range therebetween.

[0057] In some embodiments, the generate images step 950 includes generating a sequence of images. The sequence can be a sequence over time, for example, a series of X-ray images generated at different times to observe changes in the object. Examples of such changes can include perfusion of contrast agents through human surface organs. Alternatively, the sequence can be a sequence of images highlighting different X-ray energies and / or phase shifts.

[0058] In an optional smoothing step 960, one or more smoothing functions are applied to the X-ray data generated in the detect X-rays step 930. The smoothing function may optionally depend on the energy of the detected X-rays and / or on the statistics of photons detected at a particular energy. For example, the amount of smoothing may depend on the number of X-ray photons detected within a particular energy bin. Different amounts of smoothing may be applied to X-ray data within one energy range relative to one or more other energy ranges. The smoothing function applied in the smoothing step 960 is optionally configured to reduce quantum speckles in the first image, while optionally depending on the energy of the received X-ray photons. The smoothing step 960 is optionally part of the generate image step 950.

[0059] In the optional identification step 970, the image generated in the generate image step 950 is used to identify the structure within the object. The structure is optionally distinguished (e.g., highlighted or faded) within the image based on the X-ray energy and / or phase shift. Some structures can be identified using a machine learning system or other computer-assisted detection. In some embodiments, the image generation logic 160 is configured to automatically select objects with specific features within the object for the user to request. For example, the user can request to automatically identify objects with calcified tissue density features and / or with cancer density features. In one example, the user can request to highlight a narrow range of energy absorbed by a specific contrast agent (such as at the k-edge of iodine) in the first image, while fading other energies. This highlights the contrast agent. The data collected in the same X-ray exposure can be manipulated to highlight the surrounding tissue in the second image. The combination of the first image and the second image optionally provides unique clinical information (both highlighting a specific structure and showing its position relative to other tissues) in a single X-ray exposure.

[0060] Several embodiments are specifically shown and / or described herein. However, it will be understood that modifications and variations are encompassed by the above teachings and are within the scope of the appended claims without departing from the spirit and intended scope of the appended claims. For example, the energy-resolving detector described herein can be used as a nuclear camera to detect radioactive tracers and / or contrast agents within an object or material. In this case, an image can be generated by comparing a detection signal at an energy emitted by the tracer or absorbed by the contrast agent with a detection signal at a different energy. At these different energies, the X-ray imaging system 100 can generate two images (data obtained from simultaneous / X-ray exposures), one image with enhancement of the tracer / contrast agent and the other image without enhancement of the tracer / contrast agent. These images can be subtracted from each other using image generation logic 160 to generate an enhanced image.

[0061] The embodiments discussed here are explanations of the present invention. Since the reference descriptions describe these embodiments of the present invention, different modifications or adaptations to the described methods and / or specific structures can become apparent to those skilled in the art. All such modifications, adaptations or variations that rely on the teachings of the present invention and that these teachings make the advancement of the art through them are considered to be within the spirit and scope of the present invention. Therefore, these descriptions and accompanying drawings should not be considered to have a restrictive meaning, because it should be understood that the present invention is by no means limited to the embodiments shown.

[0062] The computing systems and / or logic mentioned herein may include integrated circuits, microprocessors, personal computers, servers, distributed computing systems, communication devices, network devices, etc., and different combinations thereof. The computing system or logic may also include volatile and / or non-volatile memory, such as random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), magnetic media, optical media, nano media, hard disk drives, optical disks, digital versatile disks (DVDs), optical paths, and / or other devices configured to store analog or digital information, such as stored in a database. Computer-readable media as used herein explicitly exclude paper. The computer-implemented steps of the methods indicated herein may include instruction sets stored on computer-readable media, which, when executed, cause the computing system to perform these steps. A computing system that is programmed to perform specific functions according to instructions from program software is a dedicated computing system for performing those specific functions. The data manipulated by the dedicated computing system when performing those specific functions is at least electronically stored in a buffer of the computing system, so that the dedicated computing system is physically changed from one state to the next state with each change in the stored data. "Logic" as discussed herein is explicitly defined as hardware, firmware or software, or any combination thereof, stored on a non-transient computer-readable medium. The logic may be implemented in an electronic and / or digital device (e.g., circuit) to produce a dedicated computing system. Any of the systems discussed herein may optionally include a microprocessor, including electronic circuits and / or optical paths, configured to perform any combination of the logic discussed herein. The method discussed herein may optionally include executing the logic by the microprocessor.

Claims

1. An imaging system, comprising: an X-ray source configured to generate X-rays having a range of energies; An X-ray detector comprising an array of pixels and configured to quantify the energy of detected X-rays, each of said pixels comprising: a scintillation crystal configured to generate ultraviolet light in response to the X-rays, the amount of the ultraviolet light being dependent on the energy of the photons of the X-rays; an ultraviolet light-sensitive photon detector configured to generate an electrical signal proportional to the amount of said ultraviolet light; and an electronic device configured to quantify the electrical signal, wherein the quantized electrical signal represents the energy of the X-ray; and Image generation logic is configured to generate at least a first image based on the position of each of the pixels within the array and the energy of the X-rays detected at each of the pixels.

2. The system of claim 1, further comprising a light coupling layer disposed between the scintillation crystal and the photon detector.

3. The system of any of the preceding claims, further comprising a contrast agent configured to absorb photons having a selected set of X-ray energies, wherein The image generation logic is configured to generate an image that emphasizes or deemphasizes photons having the selected set of X-ray frequencies.

4. A system as claimed in any one of the preceding claims, wherein: The scintillator crystal includes gadolinium aluminum gallium garnet doped with cerium.

5. The system of any of the preceding claims, further comprising a set of phase filters configured to select a phase shift of X-ray photons detected by the photon detector in order to distinguish between X-ray photons of different phases.

6. A system as claimed in any one of the preceding claims, wherein: The photon detector is configured to generate an electrical signal that is linearly proportional to the energy of the photon.

7. The system of any preceding claim, further comprising a light barrier disposed between the scintillation crystals of each pixel in the pixel array, the light barrier being configured to prevent light from propagating between the scintillation crystals.

8. A system as claimed in any one of the preceding claims, wherein: The pixel array includes at least 40,000 pixels.

9. A system as claimed in any one of the preceding claims, wherein: The photon detector comprises an optical diode.

10. A system as claimed in any preceding claim, wherein: The electronics include a front-end preamplifier and a pulse height detector.

11. A system as claimed in any one of the preceding claims, wherein: The image generation logic is configured to generate at least a second image based on photons having a different phase shift.

12. A system as claimed in any preceding claim, wherein: The image generation logic is configured to generate at least a second image based on photons having a different energy.

13. A system as claimed in any one of the preceding claims, wherein: The image generation logic is configured to generate an image that emphasizes or deemphasizes photons based on their energy.

14. A system as claimed in any preceding claim, wherein: The image generation logic is configured for a user to select which photons to highlight or deemphasize based on their energy, selecting from an energy array comprising at least 3, 4, 8, 10, 16 or 32 energy bins.

15. A system as claimed in any preceding claim, wherein: The image generation logic is configured to apply an energy-dependent smoothing function to the first image.

16. The system of any one of the preceding claims, further comprising a gantry configured to move the X-ray detector, wherein The imaging system is configured to generate CT (computed tomography) images.

17. A method for generating an X-ray image, the method comprising: generating X-rays using an X-ray source; passing the X-rays through a subject to be imaged; detecting the X-rays using an energy-resolving detector having a two-dimensional array of pixels, each of the pixels being configured to determine the energy of each received X-ray photon, so as to generate X-ray data simultaneously representing the two-dimensional position and energy of each X-ray, wherein the energy of the X-rays is classified into at least two energy bins; as well as At least a first image is generated based on the X-ray data.

18. The method of any preceding claim, further comprising detecting a phase shift of each received X-ray photon.

19. A method as claimed in any one of the preceding claims, wherein: The first image is a phase contrast image generated by fading or highlighting photons having one phase shift relative to photons having another phase shift.

20. The method of any preceding claim, further comprising applying a smoothing function to the X-ray data, the smoothing function being configured to reduce quantum speckle in the first image and being dependent on the energy of the received X-ray photons.

21. A method as claimed in any one of the preceding claims, wherein: The first image is an energy-resolved image in which X-ray photons having a first energy are highlighted or faded relative to photons having a second energy.

22. The method of any of the preceding claims, further comprising selecting the first energy and the second energy from at least the four energy bins.

23. The method of any of the preceding claims, further comprising generating a second image based on the X-ray data, the second image being an energy-resolved image in which photons having the second energy are highlighted or deemphasized relative to the first energy.

24. The method of any of the preceding claims, further comprising using the first image to automatically identify structures within the object, the structures being distinguished in the first image based on x-ray energy.

25. The method of any of the preceding claims, further comprising selecting photons of energy that are absorbed by a predetermined contrast agent or not absorbed by the predetermined contrast agent, and generating a second image based on highlighting or de-emphasizing or manipulating the selected photons.

Images