Method and system for image reconstruction

Abstract

A computer-implemented method of image reconstruction is disclosed, the method comprising: acquiring projection data of a region of a patient, the projection data comprising a projection representing a measured ray intensity of attenuated rays of radiation emitted from a radiation source, passing through the patient and detected at a detector, where the detector and radiation source rotate about an isocenter; performing a relative projection data correction to obtain modified projection data, the relative projection data correction including modifying a ray intensity of the projection data acquired at a substantially 180-degree offset about the isocenter; and running an image reconstruction process on the modified projection data to obtain an image of the region of the patient.

Description

Technical Field

[0001] The present invention relates to a method and system for image reconstruction. More specifically, the present invention relates to a computer-implemented method for image reconstruction using projection data associated with an object or patient, and a data processing device, a computer program and a non-transitory computer-readable storage medium configured to perform the method of image reconstruction. Background Art

[0002] Tomography is a non-invasive imaging technique that allows the internal structures of an object to be visualized without superimposing other structures that typically plague conventional projection images. For example, in a conventional chest x-ray, the ribs, heart, and lungs may be superimposed on the same film, while computed tomography (CT) slices are able to capture individual body parts in their three-dimensional positions. Tomography has been applied to a range of disciplines including medicine, physics, chemistry, and astronomy. While X-ray CT is probably the most familiar application of tomography, tomography can also be performed using alternative imaging modalities including ultrasound, magnetic resonance, nuclear medicine, and microwave technology.

[0003] Mathematically, the raw data acquired by a tomographic detector consists of a number of "projections" of the scanned object. These projections are effectively the Radon transform of the object structure. Reconstruction by means of tomographic reconstruction essentially involves solving the inverse Radon transform.

[0004] Many practical reconstruction algorithms have been developed to achieve the process of reconstructing a three-dimensional object from its projection. These algorithms are mainly designed based on the mathematics of Radon transforms, statistical knowledge of the data acquisition process, and the geometry of the data calculation method. Example reconstruction algorithms include filtered back projection algorithms and Fourier domain reconstruction algorithms.

[0005] In transmission CT modalities such as cone-beam CT (CBCT) and fan-beam CT, the presence of scatter in the projection data acquired for tomographic reconstruction is currently a key limiting factor as it causes a significant degradation in the reconstructed image quality and accuracy.

[0006] In the art, prominent approaches have been introduced to address scatter correction, including: Monte Carlo techniques, which are accurate but slow; analytical techniques, which generally lack the sophistication to accurately handle heterogeneous objects (e.g., patients); and artificial intelligence (AI) scatter correction techniques, which are hampered by computationally intensive training requirements.

[0007] In the context of medical imaging and therapy, scatter represents a bottleneck problem in volumetric image-guided therapy and adaptive radiation therapy. That is, although existing techniques produce promising results in scatter correction, these techniques are generally limited to situations where repeated scans and registrations are required and rely on the assumption that the patient configuration (e.g., positioning and weight) does not change during the course of treatment. In the context of medical imaging, high image quality (e.g., high contrast, uniformity, and resolution) during radiation therapy is an important aid in accurately adjusting, for example, cancer therapy according to the patient's current medical state. For adaptive radiation therapy, it is then desirable to utilize scanners with a large imaging field of view (FOV). In this way, changes in the patient configuration during the imaging and / or treatment process can be captured.

[0008] The inventors have recognized that there are imaging artifacts that affect the quality of reconstructed images, which are particularly (but not exclusively) prevalent in large FOV (LFOV) reconstructed images. Therefore, additional techniques are needed to correct scatter-induced artifacts in reconstructed images. Summary of the invention

[0009] The present invention is defined in the independent claim, to which reference will now be made. Further characteristics are set out in the dependent claims.

[0010] According to one aspect of the present invention, a computer-implemented method of image reconstruction is provided to reduce image artifacts. The method includes acquiring projection data of a region of a patient. The projection data may be acquired or received from an external computing device (e.g., a data storage server) or directly from an imaging or treatment system. The projection data includes a projection representing a measured ray intensity corresponding to an attenuated radiation ray. The beam of rays is emitted from a certain radiation source, passes through the region of the patient, and is subsequently detected at a detector. The detector and the radiation source may be configured on or in a gantry and may be rotated around the axis of the patient. Again, this rotation is around a radiation isocenter, which is a point in space (or the center of mass of a small bounding area) through which rays intersect when the detector and radiation source are rotated during beam on. The technology herein is also applicable to static detectors and static radiation sources arranged circumferentially around the axis of the patient. Similarly, the technology herein is also applicable to a combination of a static detector and a rotating radiation source, or a combination of a rotating detector and a static radiation source.

[0011] The method includes performing relative projection data correction. The correction process involves modifying the ray intensity of the projection data of the relative ray, which is acquired at a position approximately offset by 180 degrees around the isocenter. That is, the method modifies the ray intensity of the projection data at a first position and the ray intensity of the projection data at a second position, which is relative to the first position (180 degree offset). Of course, due to minor differences in the gantry construction, for example, the offset may not be exactly 180 degrees. For example, in the case where the offset is not exactly 180 degrees, interpolation or extrapolation of the data can be performed to ensure effective ray matching. The relative projection data correction process, which can be referred to as relative ray matching, produces modified projection data.

[0012] In this way, the projection data can be modified to address or reduce the contribution of image artifacts caused by scattering. Due to scattering asymmetry, artifacts are believed to appear in the reconstructed image. In an idealized scenario where rays passing from a first position to a second position are subjected to the same scattering process as rays passing from the second position to the first position, a relative projection data correction process would not be necessary. However, in realistic scenarios, scattering is asymmetric. This asymmetric scattering may be caused by asymmetric differences in the imaging target, which are caused by the patient structure. This asymmetric scattering may be caused by the unavoidable contribution of the treatment examination table used to support prone patients during imaging and treatment. And this asymmetric scattering may be caused by changes in source radiation intensity and detector efficiency at different positions around the table. LFOV images are particularly prone to asymmetric scattering artifacts due to the effective amplification of the asymmetry when the solid angle is expanded across the imaging area.

[0013] The method ends with an image reconstruction process performed on the modified projection data.Any conventional image reconstruction technique using projection data may be used in order to obtain a reconstructed image of the region of the patient.

[0014] The relative projection data correction process is a simple method of matching relative rays, which routinely reduces the contribution of asymmetric scatter in the reconstructed image. The relative projection data correction process is a safe technique in the sense that the technique modifies the projection data only in cases where the relative rays are different. In an idealized scenario where the scatter of the path from a first point to a second point is the same as the scatter of the path from the second point to the first point, the technique can still be applied without worrying about unnecessary data modification. Therefore, the applicability of the correction process is wide, and the technique can be simply integrated in existing image reconstruction methods and can be easily combined with other artifact correction means.

[0015] Optionally, the projection data may be fan beam computed tomography or cone beam computed tomography (CBCT) data. These imaging modalities are commonly used for medical imaging. However, the imaging modality should be an imaging modality in which projection data can be acquired from relative positions. Thus, optionally, the projection data may be positron emission tomography (PET) data or single photon emission computed tomography (SPECT) data.

[0016] Optionally, the image reconstruction method may include additional image enhancement or scatter reduction techniques. For example, the image reconstruction method may perform glare deconvolution on the projection data, which involves deconvolution of the glare spread function from the projection data (in combination with correction for so-called phantom scatter), and has been found to be useful for minimizing artifacts. The reader is referred to the work of Poludniowski, G. et al. (2011 Phys. Med. Biol. 56 1837) for additional information. Preferably, other scatter reduction techniques are performed before the relative projection data correction to ensure that the projection match is not lost or accidentally misplaced.

[0017] Optionally, the measured radiation intensity is a function of intensity signals acquired at detector elements of the detector. That is, the detector comprises a plurality of detector elements (e.g. individual pixels) and the measured radiation intensity takes into account the individual signals. In this way, the influence of random noise is reduced. For example, the measured radiation intensity may be an average of the intensities of the underlying individual detector elements.

[0018] Optionally, the image reconstruction method can be performed inline. That is, the method can correct the projection data while the imaging system acquires the projection data (rather than retrospective execution of the relative projection data performed on the complete projection data set). After the detector and the radiation source are rotated (approximately 180 degrees) to the second orientation and the projection data are acquired in the second orientation, the first ray intensity acquired for the projection when the detector and the radiation source are positioned in the first orientation undergoes relative projection data correction. The projection data can be stored in a memory or buffer until relative projection data correction is needed. This enables scatter correction to be performed continuously and enables rapid image reconstruction. Inline relative projection data correction is particularly advantageous for adaptive radiation therapy.

[0019] Alternatively, the method of image reconstruction may be applied to adaptive radiation therapy, which enables a patient's treatment to be altered or adapted in response to additional information known about the patient or signals that the patient has changed from the original state at the time of planning.

[0020] As described above, relative projection data correction is not limited to application to any particular image reconstruction technique. Rather, any image reconstruction technique (or combination thereof) that is capable of reconstructing an image of a patient from projection data is suitable. Optionally, the image reconstruction technique may involve the known Feldkamp, ​​Davis, and Kress (FDK) reconstruction method. Optionally, the reconstruction technique may involve an iterative reconstruction method; broadly speaking, these techniques begin with a hypothetical image, calculate projections from the image, compare the original projection data, and update the image based on the difference between the calculated projection and the actual projection. Advantages of iterative methods include improved insensitivity to noise and the ability to reconstruct optimal images in the presence of incomplete data.

[0021] Optionally, the reconstruction technique may involve Polyquant reconstruction, which can be considered a specific iterative reconstruction method. The reader is referred to the work of Mason, J. et al. (arXiv:1706.08743, arXiv:1805.04411) for additional information. Polyquant involves direct inference of some quantities of interest while implementing beam hardening (nonlinearity due to sources having different energies) corrections.

[0022] Optionally, the modification of the ray intensity may include modifying the first ray intensity for the projection of the first orientation (of the detector and the radiation source) using a second ray intensity. The second ray intensity may be measured for the projection of the second orientation (of the detector and the radiation source), wherein the second orientation is approximately 180 degrees offset from the first orientation about the isocenter.

[0023] Optionally, the modification of the ray intensity can include modifying the first ray intensity for the projection of the first orientation using a first measured isocenter ray intensity and using a second measured isocenter ray intensity. The isocenter ray intensity corresponds to (or is associated with) a ray from the beam that passes through the isocenter respectively. The second ray (associated with the second isocenter ray intensity) is measured at a second orientation, which is approximately 180 degrees offset from the first orientation about the center. The isocenter ray is not necessarily a ray within the beam that is located in the geometric center of the beam. The first isocenter ray and the second isocenter ray are collinear. In fact, the technique uses the isocenter ray as a metric to measure the mismatch between two relative beams, and the correction is applied to the entire beam.

[0024] Optionally, where the isocentric ray intensity is used for modification of the ray intensity, the modification may include a perturbation as set forth in equation 1 (ie, a perturbation of the projection value to the modified projection value).

[0025] CurrentProjection:=CurrentProjection+max(OpposingRay,CentralRay)-CentralRay (1)

[0026] Here, CurrentProjection is the intensity of the first ray projected at the first orientation. CentralRay is the intensity of the first isocentral ray. And OpposingRay is the intensity of the second isocentral ray.

[0027] Optionally, relative projection data correction includes modification of ray intensities and modification of scatter estimates of the projection data. That is, the relative projection data correction process can take into account ray intensities and scatter estimates. In this sense, the projection data can include measured intensities and optionally estimates of scatter or offset and gain images. For example, instead of performing relative ray matching with respect to ray intensities, relative projection data correction can perform relative ray matching with respect to ray intensities and scatter for each ray. This is particularly applicable to situations where techniques such as Polyquant are used in subsequent image reconstruction processes: scatter estimates can typically not be "subtracted" from the measured projections, but rather "compensated" during iterative updates of the reconstructed images. That is, relative ray matching can match relative ray intensities and relative scatter estimates.

[0028] Optionally, the image reconstruction method is performed for multiple projections acquired from different angles. That is, the method may include acquiring projection data and performing relative projection data correction for multiple orientations of the detector and the radiation source. In this way, asymmetric scattering-induced image artifacts may be processed for all rotation angles and the final image quality may be improved.

[0029] Optionally, the image reconstruction method may take into account curvature within (or in) the gantry (and specifically, the radiation source and detector). The position of the detector for each gantry angle may vary due to the curvature, and such curvature may be recorded by a calibration process. The relative projection data correction process may involve determining the position of each detector (or its detector element) hit by each corresponding isocenter ray (from the corresponding radiation source). By taking into account the curvature, the image reconstruction method may modify the projection data to take into account lateral, vertical and / or anterior / posterior detector offsets (e.g., from a calibrated expected position) and / or may take into account lateral, vertical and / or anterior / posterior radiation source offsets. Similarly, the image reconstruction method may take into account tilt (pitch, yaw and / or roll) of the detector and / or radiation source.

[0030] Embodiments of another aspect include a data processing apparatus including a memory storing computer-readable instructions and a processor. The processor (or controller circuit) is configured to execute the instructions to perform a computer-implemented method of image reconstruction.

[0031] Embodiments of another aspect include a computer program comprising instructions which, when executed by a computer, cause the computer to perform a computer-implemented method of image reconstruction.

[0032] Embodiments of another aspect include a non-transitory computer-readable storage medium including instructions that, when executed by a computer, cause the computer to perform a computer-implemented method of image reconstruction.

[0033] The present invention can be implemented in digital electronic circuitry, or in computer hardware, firmware, software, or a combination thereof. The present invention can be implemented as a computer program or computer program product, i.e., a computer program tangibly embodied in a non-transitory information carrier (e.g., in a machine-readable storage device or in a propagated signal) so as to be executed by one or more hardware modules or to control the operation of one or more hardware modules.

[0034] A computer program may be in the form of a stand-alone program, a computer program part, or more than one computer program, and may be written in any form of programming language (including compiled or interpreted languages), and it may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a data processing environment.

[0035] The present invention has been described with reference to specific embodiments. Other embodiments are also within the scope of the appended claims. For example, the steps of the present invention can be performed in a different order and still achieve the desired results.

[0036] Elements of the present invention have been described using the terms "processor", "input device", etc. Those skilled in the art will appreciate that such functional terms and their equivalents may refer to parts of a system that are spatially separated but combined to provide the defined functionality. Likewise, the same physical part of a system may provide two or more defined functions. For example, the same memory and / or processor may be used to appropriately implement separately defined means. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Reference is made by way of example only to the accompanying drawings in which:

[0038] Figure 1 is a flowchart of an image reconstruction method according to an embodiment;

[0039] Figure 2 is a schematic diagram of a geometric arrangement for fan beam projection or cone beam projection scanning and reconstruction;

[0040] Figure 3 is a schematic diagram of a cone beam CT scanner;

[0041] Figure 4is a schematic diagram of a geometric arrangement for relative projection data correction applied to fan beam or cone beam CT projection data;

[0042] Figure 5 is a set of reconstructed images reconstructed with and without relative projection data correction on a torso phantom;

[0043] Figure 6 is a set of reconstructed images reconstructed with and without relative projection data correction of the first clinical data;

[0044] Figure 7 is a set of reconstructed images reconstructed with and without relative projection data correction of the second, third, and fourth clinical data;

[0045] Figure 8 is a radiotherapy system adapted to implement image reconstruction according to an embodiment; and

[0046] Fig. 9 A radiotherapy device or apparatus according to an embodiment is suitable for implementing image reconstruction. DETAILED DESCRIPTION

[0047] Treatment planning typically involves performing an optimization process to optimize multiple parameters with the goal of providing a sufficiently high dose to the target region (or planning target volume, PTV) while reducing the dose to surrounding healthy tissue. A clinically appropriate plan can be obtained by balancing different treatment planning goals, for example by prioritizing treatment planning goals related to dose coverage at the target region and then prioritizing treatment plans related to protecting surrounding regions. However, the user may request unrealistically low defined dose values ​​in the surrounding regions, which negatively affects the optimization process and, in turn, the treatment planning process.

[0048] As an example, a CT image can be reconstructed from multiple projections acquired as an X-ray source rotates around an object (patient). The acquisition geometry can be defined by an acquisition FOV, which can be determined by a beam angle and can determine the maximum possible size of the reconstructed image. For a head CT scan, the FOV can be approximately 250 mm. For example, a typical LFOV commonly used for whole-body scans can have a diameter of approximately 500 mm. More specifically, systems to which the technology herein is particularly applicable have a FOV diameter of approximately 510 mm to 520 mm. A medium FOV (MFOV) system can have a FOV with a diameter of approximately 400 mm. More specifically, systems to which the technology herein is particularly applicable have a FOV diameter of approximately 410 mm to 420 mm. The inventors have recognized that due to the prevalence and increased impact of image artifacts, users typically avoid using LFOV and MFOV configurations when scanning.

[0049] Figure 1 An overall approach for image reconstruction is presented that includes relative ray matching and addresses asymmetric scatter-induced image artifacts that are particularly prevalent in LFOV systems and medium FOV (MFOV) systems.

[0050] At step 102, a computer acquires projection data of a patient. The projection data includes projections of measured ray intensities representing attenuated radiation rays. The rays are emitted from a radiation source, pass through the patient and are detected at a detector. The detector and radiation source rotate around the isocenter of the system.

[0051] At step 104, the computer performs relative projection data correction (or relative ray matching) to obtain modified projection data. The relative projection data correction process includes modifying the ray intensity of the projection data acquired at approximately 180 degree offsets around the isocenter. The relative projection data process can be considered as a data pre-processing step performed on the projection data prior to image reconstruction.

[0052] The computer runs an image reconstruction process of the modified projection data at step 106. Next, the computer obtains an image of the region of the patient. In the event that asymmetric scatter is present in the original projection data, the relative projection data correction process ensures that the associated artifacts are removed or at least significantly reduced in the reconstructed image.

[0053] Figure 2 The geometry of a typical fan beam or cone beam projection acquisition setup is illustrated, as may be used in an image reconstruction technique according to an embodiment. For simplicity, only the xy plane is illustrated; readers in the art will appreciate that, particularly for CBCT imaging, the beam also extends to a z-axis orthogonal to the xy plane. In this example, the acquisition setup captures projection data of an object 202.

[0054] The X-ray source 204 generates X-rays 206 and emits the X-rays 206 toward the object 202. In fan beam CT and CBCT, the X-rays can be considered as a beam emitted from a point source. Other imaging modalities (e.g., parallel beam systems) may include multiple sources or extended sources from which the X-rays can be emitted. The detector (or multiple detectors) 208 captures projections, which are a collection of line integrals along the path radiated from the source 204.

[0055] By rotating the source 204 and the detector 208 about the center of the image 210, multiple projections of the image can be acquired from different angles. In the present example, the source 204 and the detector 208 can be rotated in an arcuate manner along the orbital path 212. In this way, the xy plane can be rotated counterclockwise about the origin (or the center of the image 210) in a manner that maintains the relative positional relationship between the source 204 and the detector 208 while traversing the orbital path 212. Of course, other configurations of the imaging system are possible; for example, the source can be configured to rotate, and a complete detector ring can be configured to capture projections, or there can be multiple sources arranged circumferentially around a complete detector ring. Further, the imaging system can rotate the source along a spiral path so as to capture projection data along an axis of the object of interest 202 (in the present example, along the z-axis).

[0056] By processing the signals received from the detector 208, the attenuation of the intensity of the rays passing through the object 202 can be measured. By performing projection measurements at a series of different projection angles through the object 202, a sinusoidal graph can be constructed from the projection data, thereby mapping the spatial dimension of the detector array to the projection angle dimension. The intensity attenuation caused by a particular volume with an object will depict a sine wave along the spatial dimension of the detector perpendicular to the axis of rotation of the system. The volume of the object away from the center of rotation corresponds to a sine wave with a larger amplitude than the sine wave corresponding to the volume closer to the center of rotation. The phase of each sine wave in the sinusoidal graph corresponds to the relative angular position relative to the axis of rotation. By performing an image reconstruction technique (e.g., an inverse Radon transform) on the projection data in the sinusoidal graph, an image can be reconstructed, wherein the reconstructed image corresponds to a cross-sectional slice of the object 202.

[0057] The detector 208 may include underlying detector elements (e.g., pixels or pixel bins) rather than a single active area. The measured radiation intensity may then be obtained based on the underlying intensity signals acquired at the detector elements of the detector. This is advantageous because the effects of random noise, which may plague the intensity (particularly low intensities) measured at a single detector element, are reduced. For example, the measured radiation intensity may be an average (mean, mode, or median) of the underlying single detector element intensities.

[0058] Figure 3 The geometry of a fan beam or cone beam projection acquisition setup is illustrated, as incorporated into a medical CT scanner 300. A radiation source 304 emits an X-ray radiation beam that can pass through a patient supported on a viewing table within a CT scanner (not shown). A detector 308 captures the attenuated X-rays. The radiation source 304 and detector 308 can be configured to rotate within the gantry of the CT scanner to acquire projection measurements at a range of different projection angles. As shown, the angular distances between the measurement orientations need not be equal.

[0059] The image reconstruction method can take into account the curvature within the gantry (or gantry). For example, in one implementation, the curvature can be taken into account when determining where the isocenter ray hits the detector. The method can in turn determine the intensity at the isocenter location by interpolation. As an example, assume that the center of the detector at gantry angle t has coordinates (0,0); for 2D detector translation curvature, the image reconstruction method can perform the following process.

[0060] 1. The isocenter ray hits the detector at (-dx, -dy), where dx and dy are the per-projection offsets of the detector found by calibration.

[0061] 2. Linearly interpolate sub-pixel projection values ​​at (-dx, -dy).

[0062] 3. Store sub-pixel values ​​and use them for relative projection data correction.

[0063] Figure 4 is a schematic diagram of the geometric considerations for relative projection data correction applied to fan beam or cone beam CT projection data. The object of interest in this example is the patient torso. For example, the patient may lie prone on the examination table of a CT scanner, e.g. Figure 3 When the X-ray source is positioned at the first position P A When the patient is directed to the patient, it generates a beam of rays that passes through the patient's torso and is detected at a detector (not depicted) on the opposite side of the patient. The detected projection A is the signal intensity of each ray in the beam. As shown in the figure, in this example, the beam is a fan-shaped or cone-shaped beam that diverges when passing through the patient.

[0064] The radiation source and detector are then rotated 180 degrees (or its closest approximation) to the opposite side of the patient so that the source is in a second position P B . The source similarly generates a beam of rays that passes through the patient's torso and is detected at a detector on the opposite side of the patient. The detected projection B is the signal intensity of each ray within the beam. Within each beam there are collinear isocenter rays (with projections A(0) and B(0)) that pass through the isocenter of the system. The relative projection data correction process modifies the beam projections in pairs so that the ray intensity A of the first beam is modified relative to the ray intensity B of the second beam, and vice versa. For example. The ray intensities A and B can be modified or perturbed according to Equations 2 and 3:

[0065] A:=A+max(A(0),B(0))-A(0) (2)

[0066] B:=B+max(A(0),B(0))-B(0) (3)

[0067] In other words, if A(0)>B(0), then A is unchanged, and the ray intensity B is modified by a bias term equal to the difference between the isocentric ray intensities A(0) and B(0). Similarly, if B(0)>A(0), then B is unchanged, and the ray intensity A is modified by a bias term. More generally, this equation can be expressed as equation (1) set forth above.

[0068] The purpose of the ray modification according to equations (2) and (3) is to account for the difference between the projections of the relative ray orientations. This difference causes artifacts in the reconstructed image caused by scatter asymmetry. In an idealized configuration, the isocenter rays (passing through the isocenter of the system) should match. The relative projection data correction technique in this article provides an effective uniform scatter correction.

[0069] In this example, a single relative position pair (of radiation source and detector) is considered. The skilled reader will appreciate that the relative projection data correction technique can be applied to all data where projection data is acquired on opposite sides of the gantry (i.e., anywhere the detector and radiation source are positioned with a substantially 180 degree offset).

[0070] The specific form of the relative projection data correction algorithm is not limited to the above equations (2) and (3). Equations (4) and (5) illustrate another example method for modifying or disturbing the ray intensities A and B:

[0071]

[0072]

[0073] After the relative projection data correction, the modified projection data is used to perform image reconstruction to obtain an image of the target object (in this example, the patient's torso). In the following example, the inventors applied the Feldkamp, ​​Davis and Kress FDK analysis algorithm to reconstruct the CT image. The FDK algorithm is essentially a filtered back projection, which applies a filter in the frequency domain before back-projecting the projection data. The reader is referred to the medical literature, where additional details of the FDK algorithm are readily available. Similarly, the relative projection data correction process can be easily integrated into other forms of image reconstruction, such as direct Fourier reconstruction techniques and iterative reconstruction techniques, such as statistical reconstruction and algebraic reconstruction.

[0074] The relative projection data correction can be supplemented with additional artifact reduction techniques. For example, the inventors have found that glare correction by glare deconvolution of the projection is particularly advantageous. Glare deconvolution involves deconvolution of the glare spread function from the projection data (potentially in combination with correction for so-called phantom scatter), and has been found to be useful for minimizing artifacts. The reader is referred to the work of Poludniowski, G. et al. (2011 Phys. Med. Biol. 56 1837) for additional information.

[0075] The inventors have found that scatter correction is another particularly advantageous complementary artifact reduction technique. Scatter correction involves the estimation and compensation of scattered X-rays (photons) that undergo interactions but still reach the detector and in turn corrupt the measurement. Scatter estimation methods can be broadly characterized as "projection-based" or "image-based".

[0076] In projection-based scatter estimation, the scatter components can be inferred directly in the measurement domain, such as uniform scatter estimation behind the patient (see, e.g., Boellaard, R. et al., DOI: 10.1016 / S0167-8140(97)00073-X), convolution kernels (see, e.g., Ohnesorge, B. et al., DOI: 10.1007 / S003300050710), or deep learning methods (see, e.g., Maier, J. et al., DOI: 10.1007 / s10921-018-0507-z).

[0077] In image-based scatter estimation, physical simulations can be used to calculate the expected scatter using, for example, Monte Carlo codes (see, for example, Xu, Y. et al., DOI: 10.1088 / 0031-9155 / 60 / 9 / 3567).

[0078] In some cases, the scatter estimate may be subtracted from the measurement, and in other cases, the scatter estimate may be considered in the statistical model during model-based iterative reconstruction (see, eg, Erdogan, H. and Fessler, JA, DOI: 10.1109 / 42.802758).

[0079] Preferably, supplemental artifact reduction techniques (eg, scatter correction) are applied prior to relative projection data correction so that any supplemental artifact reduction techniques do not alter the projections in a manner that might result in misalignment of rays that are substantially 180 degrees offset from each other.

[0080] In one implementation, the inventors used the programming language and computing environment MATLAB to implement the above-mentioned image reconstruction method, including relative projection data correction. In a second implementation, the inventors used the programming language C++, and in a third implementation, the inventors used C++ in conjunction with a parallel computing platform and API, CUDA. Parallel processing can then be used to perform relative projection data correction. In an implementation, the projection data is stored in a GPU memory and is asynchronously updated as needed.

[0081] Figure 5 The effect of the relative projection data process on example projection data is shown. In this example, the object of interest is a torso phantom. The image is reconstructed from the projection data using the FDK algorithm. Image A illustrates a reconstructed image slice of the projection data acquired in the transverse plane of the phantom without using the relative projection data correction process. The asymmetric scattering artifact (caused by the scattering asymmetry in the relative ray projection) is clearly visible as very close bright and dark spots in the center of the image. Image B illustrates the torso phantom in the sagittal plane of the phantom without using the relative projection data correction process. The asymmetric scattering artifact takes the form of a rod-shaped artifact spanning the long axis of the phantom (due to the images reconstructed from the transverse slices, the individual transverse slices have asymmetric scattering artifacts).

[0082] Figure C illustrates a reconstructed image slice of the projection data acquired in the transverse plane of the phantom. Here, the reconstruction process uses the projection data modified using the relative projection data correction process. Figure D illustrates the torso phantom in the sagittal plane of the phantom with the relative projection data correction process. Clearly, the relative projection data correction process significantly reduces the effects of asymmetric scatter in the projection data with minimal impact on the rest of the image.

[0083] Figure 6 The effect of the relative projection data process on example projection data from a clinical patient is shown. Again, images were reconstructed from the projection data using the FDK algorithm, with Panel A illustrating a reconstructed transverse image of the patient's torso and Panel B illustrating a reconstructed sagittal slice of the patient's torso, both without the relative projection data correction process. Although not as obvious as in the previous phantom example, asymmetric scatter artifacts appear as bright spots and bright rods in the transverse and sagittal slices, respectively.

[0084] Figure C illustrates a reconstructed transverse image of the patient's torso, and Figure D illustrates a reconstructed sagittal slice of the patient's torso, both using the relative projection data correction process. As with the previous phantom example, the relative projection data correction process significantly mitigates the effects of asymmetric scatter in the projection data. Note that in the transverse reconstructed slice, there are banding artifacts that span the horizontal direction; these artifacts are likely underscanning artifacts, which occur when the angular arc spanned by the detector and radiation source is less than 360 degrees.

[0085] Figure 7 The effect of the relative projection data process on additional example clinical projection data is shown. Pictures A, C, and E show reconstructed transverse images for three different patients without the relative projection data correction process. Pictures B, C, and F show reconstructed transverse images of the same three patients with the relative projection data correction process applied. Again, in each case, the relative projection data correction process significantly mitigated the effects of asymmetric scatter in the projection data.

[0086] Figure 8 A radiotherapy apparatus suitable for acquiring projection data for image reconstruction according to an embodiment is described. A cross section of a radiotherapy apparatus 800 includes a radiation head 810 and a beam receiving device (detector) 802, both attached to a gantry 804. The radiation head 810 includes a radiation source 812 that emits a radiation beam 806. The radiation head 810 also includes a beam shaping device 818 that controls the size and shape of the radiation field associated with the beam.

[0087] The beam receiving device 802 is configured to receive radiation emitted from the radiation head 810 in order to absorb and / or measure the radiation beam. In the illustrated view, the radiation head 810 and the beam receiving device 802 are positioned diametrically opposite each other.

[0088] The gantry 804 is rotatable and supports the radiation head 810 and the beam receiving device 802 so that they can rotate about an axis of rotation 808, which can coincide with the longitudinal axis of the patient. The gantry provides for rotation of the radiation head 810 and the beam receiving device 802 in a plane perpendicular to the longitudinal axis of the patient (e.g., a sagittal plane). Three gantry directions x can be defined: G ,y G 、z G , so that y G Direction perpendicular to the gantry rotation axis. G The direction extends from the point on the gantry corresponding to the radiation head 810 toward the rotation axis of the gantry. Therefore, from the patient reference frame, y G The direction rotates as the gantry rotates.

[0089] The radiation therapy apparatus 800 also includes a support surface or examination table 820 on which a subject (or patient) is supported during radiation therapy or image acquisition. The radiation head 810 is configured to rotate about the rotation axis 808 so that the radiation head 810 directs radiation to the subject from various angles around the subject so as to spread the radiation dose received by healthy tissue to a larger area of ​​healthy tissue while accumulating a prescribed radiation dose at the target area.

[0090] The radiation therapy apparatus 800 is configured to deliver a radiation beam toward a radiation isocenter that is approximately located on a rotation axis 808 at the center of the gantry 804 , regardless of the angle at which the radiation head 810 is positioned.

[0091] The rotatable gantry 804 and radiation head 810 are sized to allow for a central aperture 822. Central aperture 822 provides an opening sufficient to allow a subject to be positioned therethrough without being accidentally contacted by the radiation head 810 or other mechanical components as the gantry rotates the radiation head 810 about the subject.

[0092] The radiation head 810 emits a radiation beam 806 along a beam axis 824 (or radiation axis or beam path), wherein the beam axis 824 is used to define the direction in which the radiation head 810 emits radiation. The radiation beam 806 is incident on a beam receiving device 802, which may include at least one of a beam stopper and a radiation detector. The beam receiving device 802 is attached to the gantry 804 on a side diametrically opposite to the radiation head 810 to attenuate and / or detect the radiation beam after the beam passes through the subject.

[0093] For example, radiation beam axis 824 may be defined as the center or point of maximum intensity of radiation beam 806 .

[0094] The beam shaper 818 defines the spread of the radiation beam 806. The beam shaper 818 is configured to adjust the shape and / or size of the radiation field generated by the radiation source. The beam shaper 818 achieves this by defining a variable shaped aperture (also referred to as a window or opening) to calibrate the radiation beam 806 to a selected cross-sectional shape. In this example, the beam shaper 818 can be provided by a combination of an aperture and an MLC. The beam shaper 818 can also be referred to as a beam modifier.

[0095] The radiation therapy device 800 can be configured to implement radiation therapy in coplanar and non-coplanar (also called tilted) modes. In coplanar therapy, radiation is emitted in a plane that is perpendicular to the axis of rotation of the radiation head 810. In non-coplanar therapy, radiation is emitted at an angle that is not perpendicular to the axis of rotation. To implement coplanar and non-coplanar therapy, the radiation head 810 can be moved between at least two positions, one in which radiation is emitted in a plane that is perpendicular to the axis of rotation (coplanar configuration) and one in which radiation is emitted in a plane that is not perpendicular to the axis of rotation (non-coplanar configuration).

[0096] In a coplanar configuration, the radiation head 810 is positioned to rotate about an axis of rotation and in a first plane. In a non-coplanar configuration, the radiation head is tilted relative to the first plane so that the radiation field generated by the radiation head is directed at an oblique angle relative to the first plane and the axis of rotation. In a non-coplanar configuration, the radiation head 810 is positioned to rotate in a corresponding second plane that is parallel to and displaced from the first plane. The radiation beam is emitted at an oblique angle relative to the second plane so that the beam sweeps out a cone when the radiation head rotates.

[0097] In one configuration, the beam receiving device 802 can remain in the same position relative to the rotatable gantry when the radiotherapy apparatus is in coplanar mode and non-coplanar mode. Thus, the beam receiving device 802 is configured to rotate about the rotation axis in the same plane in both coplanar mode and non-coplanar mode. This can be the same plane as the plane in which the radiation head rotates. In an alternative configuration, the beam receiving device 801 can also rotate.

[0098] The beam shaping device 810 is configured to reduce the spread of the radiation field in a non-coplanar configuration compared to a coplanar configuration.

[0099] Radiation therapy apparatus 800 includes a controller 830 programmed to control radiation source 812, beam receiving device 806, and gantry 802. Controller 830 may perform functions or operations such as treatment planning, treatment delivery, image acquisition, image processing, motion tracking, motion management, and / or other tasks involved in a radiation therapy procedure.

[0100] The controller 830 is programmed to control the various components of the apparatus 800, such as the gantry 804, the radiation head 810, the beam receiving device 802, and the support surface 820, in order to acquire projection data suitable for image reconstruction.

[0101] The hardware components of the controller 830 may include one or more computers (e.g., general-purpose computers, workstations, servers, terminals, portable / mobile devices, etc.); processors (e.g., central processing units (CPUs), graphics processing units (GPUs), microprocessors, digital signal processors (DSPs), field programmable gate arrays (FPGAs), dedicated or specially designed processors, etc.); memory / storage devices, such as memory (e.g., read-only memory (ROM), random access memory (RAM), flash memory, hard disk drives, optical disks, solid-state drives (SSDs), etc.); input devices (e.g., keyboards, mice, touch screens, MICS, buttons, knobs, trackballs, joysticks, handles, joysticks, etc.); output devices (e.g., displays, printers, speakers, vibration devices, etc.); circuits; printed circuit boards (PCBs); or other suitable hardware. The software components of the controller 830 may include operating device software, application software, etc.

[0102] The radiation head 810 may be connected to a head actuator 814 that is configured to actuate the radiation head 810, for example, between a coplanar configuration and one or more non-coplanar configurations, or to actuate the radiation source 812 and / or the detector 802, for example, in response to detection of a bend. This may involve translation and rotation of the radiation head 810 relative to the gantry. In some implementations, the head actuator may include a curved track along which the radiation head 810 may move to adjust the position and angle of the radiation head 810. The controller 830 may control the configuration of the radiation head 830 via the head actuator 814.

[0103] The beam shaping device 818 includes a shaping actuator 816. The shaping actuator is configured to control the position of one or more elements in the beam shaping device 818 so as to shape the radiation beam 806. In some implementations, the beam shaping device 816 includes an MLC, and the shaping actuator 816 includes a means for actuating leaves of the MLC. The beam shaping device 818 may also include an aperture, and the shaping actuator 816 may include a means for actuating a block of the aperture. The controller 830 may control the beam shaping device 818 via the shaping actuator 816.

[0104] Fig. 9 is a block diagram of an implementation of a radiation therapy system 900 suitable for performing a method of image reconstruction according to an embodiment. The example radiation therapy system 900 includes a computing system 910 within which a set of instructions for causing the computing system 910 to perform the methods (or steps thereof) discussed herein may be executed. The computing system 910 may implement an image reconstruction system. The computing system 910 may also be referred to as a computer. In particular, the methods described herein may be implemented by a processor or controller circuit 911 of the treatment planning system 910.

[0105] Computing system 910 should be considered to include any number of machines or machine collections, such as one or more computing devices, which individually or jointly execute one or more instruction sets to perform any one or more methods discussed herein. That is, hardware and / or software can be provided in a single computing device, or distributed on multiple computing devices in a computing system. In some implementations, one or more elements of a computing system can be connected (e.g., networked) to other machines in, for example, a local area network (LAN), an intranet, an extranet, or the Internet. One or more elements of a computing system can operate with the ability of a server or client machine in a client-server network environment, or operate as a peer machine in a peer-to-peer (or distributed) network environment. One or more elements of a computing system can be a personal computer (PC), a tablet computer, a set-top box (STB), a personal digital assistant (PDA), a cellular phone, a network appliance, a server, a network router, a switch or a bridge, or any machine capable of executing a set of instructions (sequential or otherwise), which specifies the action to be taken by the machine.

[0106] The computing system 910 includes a controller circuit 911 and a memory 913 (e.g., a read-only memory (ROM), a flash memory, a dynamic random access memory (DRAM) such as a synchronous DRAM (SDRAM) or a Rambus DRAM (RDRAM), etc.). The memory 913 may include a static memory (e.g., a flash memory, a static random access memory (SRAM), etc.) and / or an auxiliary memory (e.g., a data storage device), which communicate with each other via a bus (not shown). The memory 913 may be used to store or buffer projection data until relative projection data correction is required (e.g., until the radiation source and the detector have been rotated approximately 180 degrees and relative radiation data has been acquired).

[0107] Controller circuit 911 represents one or more general-purpose processors, such as microprocessors, central processing units, accelerated processing units, etc. More particularly, controller circuit 911 may include complex instruction set computing (CISC) microprocessors, reduced instruction set computing (RISC) microprocessors, very long instruction words (VLIW) microprocessors, processors implementing other instruction sets, or processors implementing a combination of instruction sets. Controller circuit 911 may also include one or more special processing devices, such as application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), digital signal processors (DSPs), network processors, etc. One or more processors of the controller circuit may have a multi-core design. Controller circuit 911 is configured to execute processing logic for executing the operations and steps discussed herein.

[0108] The computing system 910 may also include a network interface circuit 915. The computing system 910 may be communicatively coupled to an input device 920 and / or an output device 930 via an input / output circuit 916. In some implementations, the input device 920 and / or the output device 930 may be elements of the computing system 910. The input device 920 may include an alphanumeric input device (e.g., a keyboard or a touch screen), a cursor control device (e.g., a mouse or a touch screen), an audio device such as a microphone, and / or a tactile input device. The output device 930 may include an audio device such as a speaker, a video display unit (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), and / or a tactile output device. In some implementations, the input device 920 and the output device 930 may be provided as a single device or separate devices.

[0109] In some implementations, the computing system 910 may include an image processing circuit 914. The image processing circuit 914 may be configured to process image data 970 (e.g., images, imaging data, projections, projection data), such as medical images obtained from one or more imaging data sources, treatment devices 950, and / or image acquisition devices 940. The image processing circuit 914 may be configured to process or pre-process the image data 970. For example, the image processing circuit 914 may convert the received image data into a specific format, size, resolution, etc. The image processing circuit 914 may be configured to perform image reconstruction. In some implementations, the image processing circuit 914 may be combined with the controller circuit 911.

[0110] In some implementations, the radiation therapy system 900 may also include an image acquisition device 940 and / or a therapy device 650. The image acquisition device 940 and the therapy device 950 may be provided as a single device. In some implementations, the therapy device 950 is configured to perform imaging, for example in addition to providing therapy and / or during therapy.

[0111] The image acquisition device 940 may be configured to perform positron emission tomography (PET), computed tomography (CT), magnetic resonance imaging (MRI), single positron emission computed tomography (SPECT), X-ray, and the like.

[0112] Image acquisition device 940 may be configured to output image data 970 that may be accessed by computing system 910. Treatment device 950 may be configured to output treatment data 960 that may be accessed by computing system 910. Treatment data 960 may be obtained from an internal data source (e.g., from memory 913) or from an external data source (e.g., treatment device 950 or an external database).

[0113] The above methods may be implemented by a computer program. The computer program may include computer code (eg, instructions) that are arranged to instruct a computer to perform one or more of the above methods. Figure 1 The steps of the described method can be performed by computer code. The steps of the above method can be performed in any suitable order. The computer program and / or code for performing this method can be provided to a device such as a computer on one or more computer-readable media, or more generally, on a computer program product. The computer-readable medium can be transient or non-transient. The one or more computer-readable media can be, for example, electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, or propagation media for data transmission (e.g., for downloading code via the Internet). Alternatively, the one or more computer-readable media can take the form of one or more physical computer-readable media, such as semiconductor or solid-state memory, magnetic tape, removable computer disk, random access memory (RAM), read-only memory (ROM), hard disk and optical disk (e.g., CD-ROM, CD-R / W or DVD). The instructions can also reside completely or at least partially in the memory 913 and / or in the controller circuit 911 during its execution by the computer system 910, and the memory 913 and the controller circuit 911 also constitute computer-readable storage media.

[0114] In implementation, the modules, components, and other features described herein may be implemented as discrete components, or integrated into the functionality of hardware components such as ASICs, FPGAs, DSPs, or similar devices.

[0115] A "hardware component" is a tangible (e.g., non-transient) physical component (e.g., a set of one or more processors) that is capable of performing certain operations and may be configured or arranged in some physical manner. A hardware component may include dedicated circuitry or logic that is permanently configured to perform certain operations. A hardware component may include a dedicated processor, such as an FPGA or ASIC. A hardware component may also include programmable logic or circuitry that is temporarily configured by software to perform certain operations.

[0116] In addition, modules and components can be implemented as firmware or functional circuits within hardware devices. Further, modules and components can be implemented in any combination of hardware devices and software components, or only in software (e.g., code stored or otherwise embodied in a machine-readable medium or transmission medium).

[0117] Unless specifically stated otherwise, as will be apparent from the following discussion, it should be understood that throughout the specification, discussions utilizing terms such as "receiving," "determining," "comparing," "implementing," "maintaining," "identifying," "obtaining," "accessing," etc. refer to the actions and processes of a computer system or similar electronic computing device that manipulates data represented as physical (electronic) quantities within the computer system's registers and memories and transforms it into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission, or display devices.

[0118] Although certain embodiments have been described, these embodiments are presented by way of example only and are not intended to limit the scope of the invention. In fact, the novel methods and devices described herein can be embodied in various other forms; in addition, various omissions, substitutions, and changes can be made to the forms of the methods and devices described herein.

Claims

1. A computer-implemented method for image reconstruction, the method comprising: acquiring projection data of a region of a patient, the projection data comprising projections of measured ray intensities representing attenuated rays of radiation emitted from a radiation source, passing through the patient, and detected at a detector, wherein the detector and the radiation source rotate about an isocenter; performing relative projection data correction to obtain modified projection data, the relative projection data correction comprising modifying ray intensities of projection data acquired at a location offset by approximately 180 degrees about the isocenter; and An image reconstruction process is run on the modified projection data to obtain an image of the region of the patient.

2. The method according to claim 1, characterized in that: The projection data is fan beam or cone beam computed tomography (CBCT) data.

3. The method according to claim 1, characterized in that The projection data is positron emission tomography (PET) data or single photon emission computed tomography (SPECT) data.

4. The method according to claim 1, characterized in that Also includes: Glare deconvolution is performed on the projection data and / or scatter correction.

5. The method according to claim 1, characterized in that The measurement radiation intensity is a function of the intensity signals acquired at the detector elements of the detector.

6. The method according to claim 1, characterized in that Acquiring projection data and performing relative projection data correction are performed inline such that after the detector and the radiation source are rotated approximately 180 degrees about the isocenter to a second orientation and projection data is acquired at the second orientation, the intensity of the projected first rays at the first orientation of the detector and the radiation source undergoes the relative projection data correction.

7. The method according to claim 1, characterized in that The method is used for adaptive radiation therapy.

8. The method according to claim 1, characterized in that The reconstruction technique is any one of the following techniques: Feldkamp, ​​Davis and Kress (FDK) reconstruction technique; iterative reconstruction technique; and Polyquant reconstruction technique.

9. The method according to claim 1, characterized in that: Modifying the ray intensity includes modifying a first ray intensity for a projection at a first orientation of the detector and the radiation source using a second ray intensity, wherein the second ray intensity is measured for a projection at a second orientation of the detector and the radiation source, the second orientation being approximately 180 degrees offset from the first orientation about the isocenter.

10. The method according to claim 1, characterized in that Modifying the ray intensity includes modifying a first ray intensity for a projection at a first orientation of the detector and the radiation source using a first isocenter ray intensity corresponding to a first ray passing through the isocenter and using a second isocenter ray intensity corresponding to a second ray passing through the isocenter, wherein the second ray is measured at a second orientation that is approximately 180 degrees offset from the first orientation about the isocenter.

11. The method according to claim 10, characterized in that The modification includes applying the following perturbations: CurrentProjection:=CurrentProjection+max(OpposingRay,CentralRay)-CentralRay, Wherein CurrentProjection is the first ray intensity of the projection for the first orientation, CentralRay is the first isocentral ray intensity, and OpposingRay is the second isocentral ray intensity.

12. The method according to claim 1, characterized in that The relative projection data correction includes modifying the ray intensities and modifying scatter estimates of projection data acquired at approximately 180 degree offsets about the isocenter.

13. The method according to claim 1, characterized in that Acquiring projection data and performing relative projection data corrections are performed for a plurality of orientations of the detector and the radiation source.

14. The method according to claim 1, characterized in that The relative projection data correction includes adjustments for any curvature of the radiation source and / or the detector.

15. A data processing apparatus comprising a memory storing computer executable instructions and a processor configured to execute the instructions to perform a method of image reconstruction, the method comprising: acquiring projection data of a region of a patient, the projection data comprising projections of measured ray intensities representing attenuated rays of radiation emitted from a radiation source, passing through the patient, and detected at a detector, wherein the detector and the radiation source rotate about an isocenter; performing relative projection data correction to obtain modified projection data, the relative projection data correction comprising modifying ray intensities of projection data acquired at a location offset by approximately 180 degrees about the isocenter; and An image reconstruction process is run on the modified projection data to obtain an image of the region of the patient.

16. A computer program comprising instructions which, when executed by a computer, cause the computer to perform a method of image reconstruction, the method comprising: acquiring projection data of a region of a patient, the projection data comprising projections of measured ray intensities representing attenuated rays of radiation emitted from a radiation source, passing through the patient, and detected at a detector, wherein the detector and the radiation source rotate about an isocenter; performing relative projection data correction to obtain modified projection data, the relative projection data correction comprising modifying ray intensities of projection data acquired at a location offset by approximately 180 degrees about the isocenter; and An image reconstruction process is run on the modified projection data to obtain an image of the region of the patient.

17. A computer readable medium comprising instructions which, when executed by a computer, cause the computer to perform a method of image reconstruction, the method comprising: acquiring projection data of a region of a patient, the projection data comprising projections of measured ray intensities representing attenuated rays of radiation emitted from a radiation source, passing through the patient, and detected at a detector, wherein the detector and the radiation source rotate about an isocenter; performing relative projection data correction to obtain modified projection data, the relative projection data correction comprising modifying ray intensities of projection data acquired at a location offset by approximately 180 degrees about the isocenter; and An image reconstruction process is run on the modified projection data to obtain an image of the region of the patient.

Images