Multi-angle scanning coded aperture X-ray diffraction tomography system and imaging method

Abstract

This application discloses a multi-angle scanning coded aperture X-ray diffraction tomography system and method. An X-ray source generates a cone-beam incident X-ray, which is then collimated by a slit collimator to form a fan-beam incident X-ray. After irradiating the object to be imaged, it produces original diffracted X-rays, which then pass through a coded aperture template to form coded diffracted X-rays. An energy-dispersive photon counting detector detects the coded diffraction detection signals corresponding to the object at multiple imaging angles. An image reconstruction module reconstructs the multi-angle coded diffraction detection signals using a precise model of the imaging system, obtaining the diffraction tomography reconstruction results of the object. Thus, the high spatial resolution advantage of the rotating scanning X-ray diffraction tomography system is combined with the accelerated acquisition advantage of coded aperture imaging, improving the spatial resolution of X-ray diffraction tomography, reducing data acquisition time, and meeting the application requirements of X-ray diffraction tomography technology in clinical medical diagnosis and material sample analysis.

Description

Technical Field

[0001] This application relates to the field of radiation imaging technology, and in particular to a multi-angle scanning coded aperture X-ray diffraction tomography system and imaging method. Background Technology

[0002] Over the past few decades, traditional X-ray attenuation imaging has been widely used in medical, security inspection, and industrial damage detection fields. X-ray attenuation imaging is highly sensitive to the atomic number and electron density properties of materials, effectively distinguishing between light and heavy materials. However, for different materials with similar elemental compositions, the X-ray attenuation signal has limited discriminative power. In 1912, Laue proposed that the crystal planes of crystalline materials could act as gratings to produce a diffraction effect on X-rays, and since then, X-ray diffraction (XRD) has received widespread attention as a tool for analyzing the molecular arrangement of materials. Commercial powder X-ray diffractometers, based on the principle of angular dispersion X-ray diffraction (ADXRD), are applied to tasks such as quantitative analysis of crystal composition. In security inspection scenarios, energy dispersion X-ray diffraction (EDXRD) security inspection machines have played an important role in the detection of flammable liquids and explosive crystalline powders. In recent years, numerous experiments have indicated the significant value of XRD in biological sample detection applications, and the material indicative properties of XRD have shown outstanding advantages in tasks such as breast cancer detection, gallstone composition analysis, and lean meat percentage analysis of meat products.

[0003] Although XRD technology possesses powerful material identification capabilities, existing commercial XRD systems do not focus on the spatial distribution information of diffraction signals, failing to maximize the utilization of XRD technology. In mature commercial powder diffractometers, the diffraction signals measured by the system originate from the entire sample area irradiated by incident X-rays, lacking spatial resolution capability. In contrast, the security inspection EDXRD system, through the use of front and rear multi-hole collimators, measures diffraction signals at different locations in space in parallel, acquiring material distribution information at different spatial locations, thus possessing preliminary spatial resolution capability. X-ray diffraction tomography (XRDT) refers to the technique of measuring the diffraction spectrum at each location within a two-dimensional tomographic plane, representing an extension of XRD technology in two-dimensional space. The security inspection EDXRD system, employing front and rear multi-hole collimators for positioning, is a rudimentary XRDT system. Due to the geometric effects of small diffraction angles, its spatial resolution is poor, typically greater than 1 cm. Furthermore, the multi-hole rear collimator used in the security inspection EDXRD system blocks a large number of Rayleigh scattered photons, resulting in low signal utilization and long acquisition time. Subsequently, coded aperture technology was introduced into EDXRD systems. Cocoded aperture XRDT (CAXRDT), based on compressed sensing data acquisition principles, replaces the multi-pinhole collimator of EDXRD with a coded aperture, improving signal utilization by an order of magnitude and reducing sampling time from hundreds of seconds to tens of seconds. However, its spatial resolution in the transmission direction remains low. On the other hand, rotating XRDT systems employ traditional pencil-beam or fan-beam CT scanning methods, achieving the highest diffraction imaging spatial resolution currently available, easily reaching the 1mm level, based on Radon transform theory. However, in pencil-beam rotating XRDT systems, pencil translation scanning motion is still required at each rotation angle, resulting in a total imaging time of several hours. Fan-beam rotating XRDT systems use slit collimators to form fan-beam illumination and add a gridded collimator. This eliminates the need for translation scanning at each rotation angle, simplifying mechanical complexity. However, due to the gridded collimator blocking scattered photons, the scattered photon reception rate is not significantly improved, and scanning still requires several hours. Long imaging time is one of the most significant problems with rotating XRDT systems.

[0004] The limitations of spatial resolution or data acquisition time in existing XRDT systems have become the main bottlenecks for the application of XRDT technology in refined sample detection tasks. Summary of the Invention

[0005] This application provides a multi-angle scanning coded aperture X-ray diffraction tomography system and imaging method to solve the problems of low spatial resolution and long data acquisition time in related technologies.

[0006] The first aspect of this application provides a multi-angle scanning coded aperture X-ray diffraction tomography imaging system, comprising: an X-ray source for generating a cone-beam incident X-ray with a continuous energy spectrum; a slit collimator disposed in the propagation path of the cone-beam incident X-ray, such that the cone-beam incident X-ray forms a fan-beam incident X-ray after passing through the slit collimator; a stage for placing an object to be imaged, disposed in the propagation path of the fan-beam incident X-ray, such that the object to be imaged generates original diffracted X-rays after being irradiated by the fan-beam incident X-rays; and a coded aperture template disposed in the propagation path of the original diffracted X-rays, such that the original diffracted X-rays generate original diffracted X-rays after being irradiated by the fan-beam incident X-rays; and a coded aperture template disposed in the propagation path of the original diffracted X-rays. The diffracted X-rays pass through the coded aperture template to form coded diffracted X-rays; a beam blocker, disposed between the stage and the coded aperture template, is used to absorb the fan-beam transmitted X-rays that penetrate the object to be imaged; an energy-dispersive photon counting detector, disposed in the propagation path of the coded diffracted X-rays, is used to detect the coded diffracted X-rays corresponding to the object to be imaged at multiple imaging angles, obtaining multi-angle coded diffraction detection signals; an image reconstruction module is used to reconstruct the multi-angle coded diffraction detection signals using a precise model of the imaging system, obtaining the diffraction tomography reconstruction result of the object to be imaged.

[0007] Optionally, in one embodiment of this application, it further includes: a mechanical motion module, which is used to control the motion of the object to be imaged to change the imaging angle of the object relative to the X-ray source and the energy dispersive photon counting detector.

[0008] Optionally, in one embodiment of this application, it further includes: a model calculation module, used to calculate the general physical model of the imaging system based on the actual calibration parameters of the imaging system to obtain the accurate model of the imaging system, wherein the general physical model of the imaging system is the physical relationship between the multi-angle coded diffraction detection signal and the diffraction tomography reconstruction result.

[0009] Optionally, in one embodiment of this application, it further includes: a calibration module, used to calibrate the detector energy response matrix, the spectral shape of the incident X-ray beam, and the geometric parameters of the imaging system to obtain the actual calibration parameters of the imaging system.

[0010] Optionally, in one embodiment of this application, the image reconstruction module is further configured to construct an iterative optimization objective function based on the accurate model of the imaging system, and obtain the diffraction tomography reconstruction result of the object to be imaged by optimizing the iterative optimization objective function.

[0011] Optionally, in one embodiment of this application, the image reconstruction module is further configured to train a deep neural network based on the accurate model of the imaging system using multi-angle scanning coded aperture X-ray diffraction tomography data, and reconstruct the multi-angle coded diffraction detection signal through the trained deep neural network to obtain the diffraction tomography reconstruction result of the object to be imaged.

[0012] A second aspect of this application provides a multi-angle scanning coded aperture X-ray diffraction tomography method. Utilizing the multi-angle scanning coded aperture X-ray diffraction tomography system described in the above embodiments, the method includes the following steps: controlling the motion of the object to be imaged, such that the object forms multiple imaging angles relative to the X-ray source and the energy-dispersive photon counting detector; acquiring multi-angle coded diffraction detection signals corresponding to the object at the multiple imaging angles; and reconstructing the multi-angle coded diffraction detection signals using an accurate model of the imaging system to obtain the diffraction tomography reconstruction result of the object to be imaged.

[0013] Optionally, in one embodiment of this application, before reconstructing the multi-angle coded diffraction detection signal using the accurate model of the imaging system, the method further includes: calculating the general physical model of the imaging system based on the actual calibration parameters of the imaging system to obtain the accurate model of the imaging system, wherein the general physical model of the imaging system is the physical relationship between the multi-angle coded diffraction detection signal and the diffraction tomography reconstruction result.

[0014] Optionally, in one embodiment of this application, the multi-angle coded diffraction detection signal is reconstructed using an accurate model of the imaging system to obtain the diffraction tomography reconstruction result of the object to be imaged, including: constructing an iterative optimization objective function based on the accurate model of the imaging system, and obtaining the diffraction tomography reconstruction result of the object to be imaged by optimizing the iterative optimization objective function.

[0015] Optionally, in one embodiment of this application, the multi-angle coded diffraction detection signal is reconstructed using an accurate model of the imaging system to obtain the diffraction tomography reconstruction result of the object to be imaged, including: training a deep neural network based on the accurate model of the imaging system using multi-angle scanning coded aperture X-ray diffraction tomography imaging data, and reconstructing the multi-angle coded diffraction detection signal using the trained deep neural network to obtain the diffraction tomography reconstruction result of the object to be imaged.

[0016] This application presents a multi-angle scanning coded aperture X-ray diffraction tomography system and imaging method, which combines the high spatial resolution advantage of rotating scanning XRDT with the accelerated acquisition advantage of coded aperture imaging. It proposes a new high spatial resolution fast XRDT system, defined as a multi-angle scanning coded aperture X-ray diffraction tomography system (multi-angle scanning CAXRDT), which can effectively meet the application requirements of XRDT technology in clinical medical diagnosis and material sample analysis.

[0017] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0018] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0019] Figure 1 This is a schematic diagram of a multi-angle scanning coded aperture X-ray diffraction tomography system provided according to an embodiment of this application;

[0020] Figure 2 A schematic diagram of the encoding hole template provided in the embodiments of this application;

[0021] Figure 3 A schematic diagram of an alcohol sample provided in the embodiments of this application;

[0022] Figure 4 A schematic diagram of alcohol sample coded diffraction detection signal provided in the embodiments of this application;

[0023] Figure 5 A schematic diagram of the reconstruction results of an alcohol sample at different scattering vectors provided in the embodiments of this application;

[0024] Figure 6 This is a flowchart of a multi-angle scanning coded aperture X-ray diffraction tomography method provided according to an embodiment of this application. Detailed Implementation

[0025] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0026] XRD signals are the macroscopic manifestation of Rayleigh scattering between X-ray photons and matter, reflecting intermolecular structural information and possessing strong material indicative properties. In recent years, the application value of XRD technology in medical and security inspections has become a hot topic, and XRDT is an inevitable technological direction for the refinement and visualization of XRD detection. However, existing security inspection EDXRD and snapshot CAXRDT systems have a spatial resolution greater than 1 cm in the transmission direction, resulting in severe volume effects when detecting samples with high-frequency spatial structures. On the other hand, rotational scanning XRDT can achieve high spatial resolution sample detection, but the scanning time is as long as several hours, which is difficult to meet the needs of practical applications. The embodiments of this application aim to combine the isotropic high spatial resolution advantage of rotational scanning methods with the rapid parallel advantage of compressed sensing data acquisition via coded apertures, proposing a new refined XRDT system suitable for practical applications with a shorter scanning time, called the multi-angle scanning coded aperture X-ray diffraction tomography system (multi-angle scanning CAXRDT), thereby meeting the requirements of clinical medical diagnosis and material sample analysis applications for imaging speed and spatial accuracy.

[0027] The following description, with reference to the accompanying drawings, describes an embodiment of the multi-angle scanning coded aperture X-ray diffraction tomography system of this application.

[0028] Figure 1 This is a schematic diagram of a multi-angle scanning coded aperture X-ray diffraction tomography system provided according to an embodiment of this application.

[0029] like Figure 1 As shown, the multi-angle scanning coded aperture X-ray diffraction tomography system includes: an X-ray source commonly used in medical or security inspection, a slit collimator, a stage for placing samples / objects, a radiation blocker, a coded aperture template, and an energy dispersive photon counting detector.

[0030] In the multi-angle scanning CAXRDT system imaging process, the X-ray source generates a conical beam of incident X-rays with a continuous energy spectrum. A slit collimator is positioned in the propagation path of the conical beam X-rays, which, after passing through the slit collimator, form a fan-beam X-ray. The stage is positioned in the propagation path of the fan-beam X-rays, and the object to be imaged is irradiated by the fan-beam X-rays, generating primary diffracted X-rays. A coding aperture template is positioned in the propagation path of the primary diffracted X-rays, which, after passing through the coding aperture template, form coded diffracted X-rays. A radiation blocker is positioned between the stage and the coding aperture template, absorbing the portion of the fan-beam X-rays that directly penetrates the object. An energy-dispersive photon counting detector is positioned in the propagation path of the coded diffracted X-rays to detect the coded diffracted X-rays corresponding to the object at multiple imaging angles, obtaining multi-angle coded diffraction detection signals. The image reconstruction module is used to reconstruct multi-angle coded diffraction detection signals using an accurate model of the imaging system, and obtain the diffraction tomography reconstruction results of the object to be imaged.

[0031] At the imaging angle Below, the coded diffraction detection signal obtained by the energy-dispersive photon counting detector is Where u,v are the pixel position coordinates of the two-dimensional detector, and E is the detector energy channel.

[0032] Optionally, in embodiments of this application, the multi-angle scanning coded aperture X-ray diffraction tomography system further includes a mechanical motion module. The mechanical motion module is used to control the movement of the object to be imaged, so as to change the imaging angle of the object relative to the X-ray source and the energy dispersive photon counting detector.

[0033] During the scanning process, the mechanical motion module controls the object to be imaged to change its imaging angle relative to the X-ray source and detector, acquiring coded diffraction X-rays at each angle, ultimately obtaining multi-angle coded diffraction detection signals. in This represents the total number of scanning angles.

[0034] The image reconstruction module utilizes a precise model of the imaging system to encode the multi-angle diffraction detection signal I. XRD The reconstruction yielded the diffraction fault reconstruction results. Where x and y are two-dimensional spatial coordinates, and q is the scattering vector.

[0035] Optionally, in one embodiment of this application, the multi-angle scanning coded aperture X-ray diffraction tomography system further includes a model calculation module. The model calculation module is used to calculate the general physical model of the imaging system based on the actual calibration parameters of the imaging system to obtain an accurate model of the imaging system, wherein the general physical model of the imaging system is the multi-angle coded diffraction detection signal I. XRDThe physical relationship between the result f and the diffraction fault reconstruction result.

[0036] The center of the field of view is defined as the origin of the global coordinate system, denoted by o. The axis perpendicular to the detector plane is taken as the X-axis, the Y-axis is parallel to the slit of the slit collimator, and the Z-axis is perpendicular to the slit and parallel to the detector plane. This represents the equivalent rotation angle of an object from different viewpoints, using the first-person viewpoint coordinate system as a reference. The general physical model of an imaging system refers to The physical relationship between f(x,y,q) and f(x) is expressed as follows:

[0037]

[0038] Where R(E,E') is the energy response matrix of the detector. The energy spectrum shape of the fan-beam incident X-rays. The incident attenuation factor refers to... At an angle, the attenuation ratio of an X-ray photon with energy E' from the light source to points x and y on the object. The diffraction attenuation factor refers to... At an angle, the attenuation ratio of an X-ray photon with energy E' from x,y on the object to u,v on the detector. Let be the encoding aperture factor, representing whether a scattered photon from position x, y on the object can pass through the aperture of the encoding aperture template to reach position u, v on the detector. Ideally, it is a binary function taking values ​​of 0 / 1, but in practice, it can take values ​​in the interval [0, 1]. Let Let be the Rayleigh scattering factor, describing the intrinsic relationship model of Rayleigh scattering when the diffraction intensity is unaffected by the coding aperture template and attenuation effect. Wherein, Represents the diffraction solid angle, describing the size of the diffraction solid angle per unit area of ​​the detector from x, y on the object to u, v on the detector. For the Thomson scattering cross section, N A is Avogadro's constant, h is Planck's constant, and c is the speed of light. The incident attenuation factor in formula (1) Diffraction attenuation factor Diffraction angle θ S and diffraction solid angle Expressed as:

[0039]

[0040]

[0041]

[0042]

[0043] In formulas (2) and (3), μ(x,y,z,E) represents the three-dimensional attenuation coefficient distribution of the object to be reconstructed at energy E, and l{(x1,y1,z1),(x2,y2,z2)} refers to the integral path from point (x1,y1,z1) to (x2,y2,z2) in the three-dimensional coordinate system. In formula (5)<a,b> This represents the angle between two vectors a and b. S is the distance from the X-ray source to the origin o, and D is the distance from o to the detector, as shown below. Figure 1 As shown, S and D can take different values ​​at different angles, thus forming a multi-angle CAXRDT scan of a non-circular orbit.

[0044] Optionally, in embodiments of this application, the multi-angle scanning coded aperture X-ray diffraction tomography system further includes: a calibration module, used to calibrate the detector energy response matrix, the spectrum of the fan-beam incident X-rays, and the geometric parameters of the imaging system to obtain the actual calibration parameters of the imaging system.

[0045] The precise model of the multi-angle scanning coded aperture X-ray diffraction tomography system is obtained by numerical calculation based on the actual calibration parameters of the general physical model of the multi-angle scanning coded aperture X-ray diffraction tomography system. This process includes the calibration of the detector energy response matrix R(E,E') and the spectral shape of the fan-beam incident X-rays. Calibration, system geometric parameter calibration, and system factor calculation.

[0046] Detector energy response matrix R(E,E') calibration: The detector energy response matrix can be obtained by calibration using metal powder X-ray fluorescence, crystal powder diffraction, or Monte Carlo simulation.

[0047] Spectral shape of fan-beam incident X-rays Calibration: The measured optomechanical spectrum was directly measured using an energy-dispersive photon counting detector, and the spectral shape of the fan-beam incident X-rays was estimated from the measured optomechanical spectrum based on the detector energy response matrix R(E,E').

[0048] System geometric parameter calibration refers to the spatial position calibration of the energy dispersive photon counting detector and the spatial position calibration of the encoding aperture template. It is also calibrated by measuring the change relationship between the direct X-ray signal measured by the detector and the motor stepping translation under X-ray beam irradiation by replacing the pen beam collimator.

[0049] System factor calculation: Based on the system geometric parameter calibration results, the calculation formula (1) is used in the case of discretized object pixels and discretized detector pixels. and The specific values ​​are used to obtain the actual accurate model of the multi-angle scanning CAXRDT, denoted as H.XRD ,Right now:

[0050] I XRD =H XRD (f) (6)

[0051] Optionally, in one embodiment of this application, the image reconstruction module is further configured to construct an iterative optimization objective function based on the accurate model of the imaging system, and to obtain the diffraction tomography reconstruction result of the object to be imaged by optimizing the iterative optimization objective function.

[0052] Multi-angle encoded diffraction detection signal I XRD Reconstructed for H XRD The process of inversion. This includes model-based iterative reconstruction methods. Based on the actual accurate model H of the multi-angle scanning CAXRDT. XRD Construct the iterative optimization objective function:

[0053]

[0054] Among them l MAP 0 represents the overall iterative optimization objective function, l fidelity 0 represents the data fidelity cost function, which is derived from the statistical noise model. XRD and H XRD (f) Signal matching quantization description, such as negative log-likelihood function, statistical noise model may include Poisson noise model, Gaussian noise model. prior 0 represents the prior cost function for the image, which may include total variation minimization, nonlocal mean filtering, dictionary learning, low rank, etc. f represents the diffraction tomography image to be optimized.

[0055] The diffraction tomography reconstruction result f can be obtained by optimizing f using the gradient descent method, the alternating iterative optimization method (ADMM), or the split Bregman method.

[0056]

[0057] Optionally, in one embodiment of this application, the image reconstruction module is further configured to train a deep neural network based on a precise model of the imaging system using multi-angle scanning coded aperture X-ray diffraction tomography data, and reconstruct the multi-angle coded diffraction detection signal through the trained deep neural network to obtain the diffraction tomography reconstruction result of the object to be imaged.

[0058] Multi-angle encoded diffraction detection signal I XRD Reconstructed for H XRD The process of finding the inverse also includes using reconstruction methods based on deep neural networks.

[0059] Design a deep neural network, denoted as . θ ReconThese are the parameters of the neural network. A large amount of multi-angle scanning CAXRDT data was obtained through simulation and actual experiments to train the deep neural network, resulting in the parameters θ of the trained neural network. * Recon .use Indicates H XRD The process of finding the inverse. The reconstruction process is then as follows:

[0060]

[0061] The multi-angle scanning coded aperture X-ray diffraction tomography system of this application can complete a scan in a few minutes, making it the first high-precision diffraction tomography system in the field with a scan time of minutes. It combines the high spatial resolution advantage of rotational scanning with the rapid parallel data acquisition advantage of coded aperture compressed sensing, achieving diffraction pattern imaging of the scanned object within minutes, acquiring the diffraction pattern at each pixel, and indicatively imaging the material distribution of the scanned object. The spatial resolution of the material distribution imaging can reach the 1mm level. The multi-angle scanning CAXRDT system effectively meets the requirements of clinical medical diagnosis and material sample analysis applications for imaging speed and spatial accuracy.

[0062] We propose an easy-to-implement system calibration method, an accurate model calculation method, and an image reconstruction method for the multi-angle scanning CAXRDT system. These methods, when used in conjunction with the multi-angle scanning CAXRDT system, fully consider physical and practical factors such as discretization implementation factors, attenuation factors, and changes in diffraction angle and diffraction solid angle. They can accurately and stably reconstruct the data acquired by the multi-angle scanning CAXRDT system and achieve optimal results.

[0063] The multi-angle scanning coded aperture X-ray diffraction tomography system of this application will be described in detail below through specific embodiments.

[0064] The multi-angle scanning CAXRDT system uses a tungsten anode X-ray source, with a tube voltage of 100kV and a current of 5mA during operation. A fan-beam X-ray is formed using two-stage slit collimators. The first-stage slit collimator has a slot length of 40mm and a width of 0.5mm, located 100mm from the X-ray source focal point. The second-stage slit collimator has a slot length of 50mm and a width of 0.5mm, located 300mm from the X-ray source focal point. The optical path from the X-ray source exit to the two-stage slit collimators is wrapped with lead sheeting to shield excess X-rays. The stage is an 80mm diameter turntable, located 400mm from the X-ray source focal point. During use, the sample placement area is a 50mm diameter circle centered on the stage's rotation center, which constitutes the system's imaging field of view. The coded aperture template and the energy-dispersive photon counting detector are mounted together on a mechanical platform that can translate along the YZ direction. The coded aperture template is 150 mm from the rotation center of the stage, and the energy-dispersive photon counting detector is 300 mm from the rotation center of the stage. Their relative positions are fixed, both located above the fan-beam incident X-ray. In this embodiment, the coded aperture template is a 1.5 mm thick tungsten plate, on which 40 columns and 8 rows of candidate aperture positions (320 candidate aperture positions) are planned and distributed in a 40 mm × 8 mm strip area. The periodic interval of the rows and columns of candidate aperture positions is 2 mm. 160 candidate aperture positions are randomly selected to process through holes. Each through hole is a square with a side length of 1 mm. A schematic diagram of the coded aperture template is shown below. Figure 2 As shown, the energy-dispersive photon counting detector has 64×16 detector pixels, with a pixel side length of 1.6 mm. In operation, its measurement energy spectrum range is set to 21 keV-100 keV, with an interval of 1 keV.

[0065] During sample scanning, the sample is placed within the imaging field of view of the system on the stage. Fifteen data acquisition angles are evenly set within a 360° circle, with a detector data acquisition time of 80 seconds for each angle. Ignoring stage movement time, the total scanning time is 20 minutes. (Since the detector in this embodiment is only located above the fan-beam incident X-ray, if the detection areas of the coded aperture template and the optomechanical counting detector simultaneously cover both sides of the fan-beam incident X-ray, the time can be halved.) The coded diffraction detection signals I at the 15 angles... XRD It is stored for subsequent reconstruction. Figure 3 and Figure 4 A schematic diagram of the coded diffraction detection signal at the 50keV energy channel of the alcohol sample at a certain angle is given in this embodiment.

[0066] Before integrating the energy-dispersive photon counting detector into the multi-angle scanning CAXRDT system, the detector's energy response matrix R(E,E') was pre-calibrated using the metal powder fluorescence method. The energy spectrum at 100 keV was measured using this energy-dispersive photon counting detector, and the spectral shape of the fan-beam incident X-ray at 100 keV was obtained by inverting the detector response.

[0067] An energy-dispersive photon counting detector was installed in a multi-angle scanning CAXRDT system. A pinhole collimator was installed in front of the X-ray source to generate a pen-beam X-ray to irradiate the detector. The detector was translated along the YZ direction in 0.1 mm steps, and the change of the pixel with the highest count rate with the translation in the YZ direction was observed to obtain the position Y of the detector relative to the central pen-beam. D = -1.7mm, Z D =2.9mm.

[0068] Further install the encoding hole template, cover the remaining encoding holes with lead sheet, leaving only the center encoding hole as the calibration hole. Adjust the YZ motor in 0.1mm translation steps to maximize the total count rate of the pencil beam X-rays passing through this calibration hole and illuminating the detector. At this point, the calibration hole is directly aligned with the center pencil beam. Record the motor translation position. C = -0.8mm, Z C =1.0mm. Therefore, the actual Y-coordinate of the encoding hole template is -Y. C 0.8mm, Z coordinate -Z C The value is -1.0mm. To remove the obstructing material on the coding hole template, replace the pinhole collimator with a slit collimator, install the X-ray blocker, and adjust the multi-angle scanning CAXRDT system back to working condition.

[0069] Based on the system calibration results above, the system factor in formula (1) is calculated to obtain the actual accurate model H of the multi-angle scanning CAXRDT. XRD .

[0070] During multi-angle scanning CAXRDT reconstruction, a model-based iterative reconstruction method is used to reconstruct the encoded diffraction probe signal I. XRD For reconstruction, the maximum a posteriori probability cost function is first constructed. The likelihood function adopts a Poisson noise distribution model. The image prior uses a total variation minimum prior in the spatial dimension and a second-order smoothing term prior in the diffraction spectrum dimension. Therefore, the maximum a posteriori probability cost function in this embodiment is:

[0071] l MAP (f,I XRD )=l Poisson (I XRD H XRD (f))+l TV-xy (f)++l Smooth-q (f) (10)

[0072] Among them l Poisson Let l be the likelihood function of the Poisson distribution. TV-xy Let l be the total variational minimum cost function in spatial dimension. Smooth-qThe cost function is the second-order smoothing constraint for the diffraction spectrum dimension. The diffraction tomography reconstruction result f is obtained by optimizing equation (10) using the split-Bregman method, as shown below. Figure 5 As shown, Figure 5 In the table, number 1 represents 80% alcohol by mass, number 2 represents 40% alcohol by mass, and number 3 represents water.

[0073] A segmentation-based method is used to estimate the three-dimensional attenuation coefficient distribution of the scanned object. For each scanned object, during the first reconstruction, the attenuation of rays by the object is not considered (μ(x,y,z,E)≡0), and a preliminary reconstruction is performed. The preliminary reconstruction result is then segmented using a threshold, and the attenuation coefficient of water is used to approximate the three-dimensional attenuation coefficient distribution of the scanned object in the areas containing the object. After obtaining the three-dimensional attenuation coefficient distribution of the scanned object, the incident attenuation factor is calculated. Diffraction attenuation factor And the encoded diffraction detection signal I XRD Accurate reconstruction yields the results of diffraction fault reconstruction.

[0074] The multi-angle scanning coded aperture X-ray diffraction tomography system proposed in this application combines the high spatial resolution advantage of rotating scanning XRDT with the accelerated acquisition advantage of coded aperture imaging, and proposes a new high spatial resolution fast XRDT system, defined as multi-angle scanning CAXRDT, which can effectively meet the application needs of XRDT technology in clinical medical diagnosis and material sample analysis.

[0075] Next, referring to the accompanying drawings, a multi-angle scanning coded aperture X-ray diffraction tomography method according to an embodiment of this application is described.

[0076] Figure 6 This is a flowchart of a multi-angle scanning coded aperture X-ray diffraction tomography method provided according to an embodiment of this application.

[0077] like Figure 6 As shown, this multi-angle scanning coded aperture X-ray diffraction tomography method utilizes the multi-angle scanning coded aperture X-ray diffraction tomography system described in the above embodiment. The imaging method includes the following steps:

[0078] Step S101: Control the movement of the object to be imaged so that the object to be imaged forms multiple imaging angles relative to the X-ray source and the energy dispersive photon counting detector.

[0079] In step S102, multi-angle coded diffraction detection signals of the object to be imaged at multiple imaging angles are acquired.

[0080] In step S103, the multi-angle coded diffraction detection signal is reconstructed using the accurate model of the imaging system to obtain the diffraction tomography reconstruction result of the object to be imaged.

[0081] Optionally, in one embodiment of this application, before reconstructing the multi-angle coded diffraction detection signal using the accurate model of the imaging system, the method further includes: calculating the general physical model of the imaging system based on the actual calibration parameters of the imaging system to obtain the accurate model of the imaging system, wherein the general physical model of the imaging system is the physical relationship between the multi-angle coded diffraction detection signal and the diffraction tomography reconstruction result.

[0082] Optionally, in one embodiment of this application, the multi-angle coded diffraction detection signal is reconstructed using an accurate model of the imaging system to obtain the diffraction tomography reconstruction result of the object to be imaged, including: constructing an iterative optimization objective function based on the accurate model of the imaging system, and obtaining the diffraction tomography reconstruction result of the object to be imaged by optimizing the iterative optimization objective function.

[0083] Optionally, in one embodiment of this application, the multi-angle coded diffraction detection signal is reconstructed using an accurate model of the imaging system to obtain the diffraction tomography reconstruction result of the object to be imaged, including: training a deep neural network based on the accurate model of the imaging system using multi-angle scanning coded aperture X-ray diffraction tomography imaging data, and reconstructing the multi-angle coded diffraction detection signal using the trained deep neural network to obtain the diffraction tomography reconstruction result of the object to be imaged.

[0084] Optionally, in one embodiment of this application, the method further includes: calibrating the detector energy response matrix, the spectral shape of the incident X-ray beam, and the geometric parameters of the imaging system to obtain the actual calibration parameters of the imaging system.

[0085] It should be noted that the foregoing explanation of the embodiment of the multi-angle scanning coded aperture X-ray diffraction tomography system also applies to the multi-angle scanning coded aperture X-ray diffraction tomography method of this embodiment, and will not be repeated here.

[0086] The multi-angle scanning coded aperture X-ray diffraction tomography method proposed in the embodiments of this application combines the high spatial resolution advantage of rotating scanning XRDT with the accelerated acquisition advantage of coded aperture imaging, and proposes a new high spatial resolution fast XRDT system, defined as multi-angle scanning CAXRDT, which can effectively meet the application requirements of XRDT technology in clinical medical diagnosis and material sample analysis.

[0087] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0088] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0089] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

Claims

1. A multi-angle scanning coded aperture X-ray diffraction tomography system, characterized in that, include: An X-ray source used to generate a cone-beam incident X-ray with a continuous energy spectrum; A slit collimator is disposed in the propagation path of the cone-beam incident X-rays so that the cone-beam incident X-rays form a fan-beam incident X-rays after passing through the slit collimator. A stage for placing the object to be imaged is positioned on the propagation path of the fan-beam incident X-rays, so that the object to be imaged produces original diffracted X-rays after being irradiated by the fan-beam incident X-rays. A coding aperture template is disposed in the propagation path of the original diffracted X-rays so that the original diffracted X-rays form coded diffracted X-rays after passing through the coding aperture template; A radiation blocker is disposed between the stage and the encoding aperture template to absorb fan-beam transmitted X-rays that penetrate the object to be imaged. An energy-dispersive photon counting detector is disposed on the propagation path of the coded diffraction X-rays and is used to detect the coded diffraction X-rays corresponding to the object to be imaged at multiple imaging angles to obtain multi-angle coded diffraction detection signals. The image reconstruction module is used to reconstruct the multi-angle coded diffraction detection signal using the accurate model of the imaging system, so as to obtain the diffraction tomography reconstruction result of the object to be imaged. A mechanical motion module is used to control the movement of the object to be imaged, so as to change the imaging angle of the object relative to the X-ray source and the energy dispersive photon counting detector.

2. The system according to claim 1, characterized in that, Also includes: The model calculation module is used to calculate the general physical model of the imaging system based on the actual calibration parameters of the imaging system to obtain the accurate model of the imaging system. The general physical model of the imaging system is the physical relationship between the multi-angle coded diffraction detection signal and the diffraction tomography reconstruction result.

3. The system according to claim 2, characterized in that, Also includes: The calibration module is used to calibrate the detector energy response matrix, the spectral shape of the incident X-rays from the fan beam, and the geometric parameters of the imaging system to obtain the actual calibration parameters of the imaging system.

4. The system according to claim 1 or 2, characterized in that, The image reconstruction module is further used to construct an iterative optimization objective function based on the accurate model of the imaging system, and to obtain the diffraction tomography reconstruction result of the object to be imaged by optimizing the iterative optimization objective function.

5. The system according to claim 1 or 2, characterized in that, The image reconstruction module is further used to train a deep neural network based on the accurate model of the imaging system using multi-angle scanning coded aperture X-ray diffraction tomography data, and to reconstruct the multi-angle coded diffraction detection signal through the trained deep neural network to obtain the diffraction tomography reconstruction result of the object to be imaged.

6. A multi-angle scanning coded aperture X-ray diffraction tomography method, utilizing the multi-angle scanning coded aperture X-ray diffraction tomography system according to any one of claims 1-5, characterized in that, Includes the following steps: The motion of the object to be imaged is controlled so that the object to be imaged forms multiple imaging angles relative to the X-ray source and the energy dispersive photon counting detector; Collect multi-angle encoded diffraction detection signals of the object to be imaged at multiple imaging angles; The multi-angle coded diffraction detection signal is reconstructed using an accurate model of the imaging system to obtain the diffraction tomography reconstruction result of the object to be imaged.

7. The method according to claim 6, characterized in that, Before reconstructing the multi-angle coded diffraction detection signal using a precise model of the imaging system, the following steps are also included: The general physical model of the imaging system is calculated based on the actual calibration parameters of the imaging system to obtain the accurate model of the imaging system. The general physical model of the imaging system is the physical relationship between the multi-angle coded diffraction detection signal and the diffraction tomography reconstruction result.

8. The method according to claim 6, characterized in that, The multi-angle coded diffraction detection signal is reconstructed using a precise model of the imaging system to obtain the diffraction tomography reconstruction result of the object to be imaged, including: An iterative optimization objective function is constructed based on the accurate model of the imaging system, and the diffraction tomography reconstruction result of the object to be imaged is obtained by optimizing the iterative optimization objective function.

9. The method according to claim 6, characterized in that, The multi-angle coded diffraction detection signal is reconstructed using a precise model of the imaging system to obtain the diffraction tomography reconstruction result of the object to be imaged, including: Based on the accurate model of the imaging system, a deep neural network is trained using multi-angle scanning coded aperture X-ray diffraction tomography data. The trained deep neural network is then used to reconstruct the multi-angle coded diffraction detection signal to obtain the diffraction tomography reconstruction result of the object to be imaged.

Images