Noise and artifact reduction for image scattering correction

Abstract

An imaging apparatus and associated method are provided to receive measurement projection data in a primary partition and to correct scattering by processing the imaging data into two separate components: non-scattering corrected data and scattering-only data. Separate image processing (e.g., reconstruction) allows for the use of personalized data processing suitable for the source data, including filters, thereby focusing on specific aspects of the source data, such as noise and artifact reduction, resolution, edge preservation, etc. Combining the separately processed data results in an optimized balance between these aspects and improved image quality.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority to U.S. Patent Application Serial No. 16 / 823,560, filed March 19, 2020, entitled "Noise and Artifact Reduction for Image Scattering Correction". This application also relates to U.S. Patent Application Serial No. 16 / 694,145, filed November 25, 2019, entitled "Multimode Radiation Apparatus and Method", and U.S. Patent Application Serial No. 16 / 694,148, filed November 25, 2019, entitled "Apparatus and Method for Scalable Field Imaging Using a Multi-Source System", the entire contents of which are incorporated herein by reference. Technical Field

[0003] The disclosed aspects of the technology relate to improving the quality during radiological image processing, including, for example, reducing noise and artifacts associated with scattering and scatter correction, and more specifically to processing scatter-corrected images into non-scatter-corrected components and scatter-only components. Background Technology

[0004] Tomography is a non-invasive radiographic imaging technique used to generate cross-sectional images of three-dimensional (3D) objects without superimposing tissue. Tomography can be classified into transmission tomography, such as computed tomography (CT), and emission tomography, such as single-photon emission computed tomography (SPECT) and positron emission tomography (PET). CT is a technique based on transmitting X-rays through a patient to produce images of body parts. SPECT and PET provide 3D imaging information about radionuclides injected into the patient's body, showing metabolic and physiological activities within organs.

[0005] In tomography, projections are acquired from many different angles around the body via one or more rotating detectors (along with the rotating radiation source in CT). These data are then reconstructed to form a 3D image of the body. For example, reconstruction of tomographic images can be achieved via filtered backprojection and iterative methods.

[0006] The quality of the final image is limited by several factors. Some of these are the attenuation and scattering of gamma-ray photons, detection efficiency, and the spatial resolution of the collimator-detector system. These factors can lead to poor spatial resolution, low contrast, and / or high noise levels. Image data processing techniques (e.g., filtering) can be used to improve image quality.

[0007] In CT, including cone-beam CT, the primary signal detected by the detector element represents the X-rays that exit the tube, penetrate the patient's body, and reach or are detected by the detector. The X-rays in the primary signal travel along the X-ray path that connects the tube's focal point to the detector element. The scattered signal detected by the same element also represents the X-rays scattered into the element. This primary signal allows for the reconstruction of CT images. However, the scattered signal can degrade CT images both quantitatively and qualitatively.

[0008] Scattering in various radiographic imaging modalities, including CT and cone-beam CT, can explain a significant portion of the detected photons. Scattering can negatively impact image quality, including contrast and quantitative accuracy. Therefore, scattering measurements, estimations, and corrections can be applied to data processing and image reconstruction, including in the case of image-guided radiation therapy (IGRT). IGRT utilizes medical imaging techniques such as CT to acquire images of patients before, during, and / or after treatment.

[0009] Compared to no scattering correction, software-based scattering correction can significantly increase noise across various radiographic imaging modalities. For example, in low-count scans (low applied dose and / or large patients), significant scattering noise associated with scattering correction at certain angles may be amplified and manifest as strong striated artifacts in reconstructed images. Therefore, noise reduction for scattering correction is a challenging task in image generation. Summary of the Invention

[0010] In one embodiment, a method for generating a radiological image includes receiving radiation data from a radiographic imaging device, wherein the radiation data includes a principal component and a scattering component; generating a non-scattering correction dataset based on the radiation data and using a first data processing technique; estimating the scattering component of the radiation data; generating a scattering-only dataset based on the scattering estimation and using a second data processing technique, wherein the second data processing technique differs from the first data processing technique; and generating an image based on the non-scattering correction dataset and the scattering-only dataset.

[0011] Features described and / or illustrated with respect to one embodiment may be used in the same or similar manner in one or more other embodiments and / or in combination with or in place of features of other embodiments.

[0012] The description of this invention does not in any way limit the scope of the terms used in the claims or the claims or invention. The terms used in the claims have their full ordinary meaning. Attached Figure Description

[0013] Embodiments of the invention are illustrated in the accompanying drawings, which are incorporated in and constitute a part of this specification, and together with the general description of the invention given above and the detailed description given below, serve to illustrate embodiments of the invention. It will be understood that the element boundaries (e.g., boxes, groups of boxes, or other shapes) shown in the drawings represent one embodiment of a boundary. In some embodiments, one element may be designed as multiple elements, or multiple elements may be designed as one element. In some embodiments, an element shown as an inner part of another element may be implemented as an outer part, and vice versa. Furthermore, elements may not be drawn to scale.

[0014] Figure 1 This is a flowchart depicting an exemplary method for generating radiographic images by treating imaging data as non-scattering correction components and scattering-only components.

[0015] Figure 2 This is a flowchart depicting another exemplary method for generating radiographic images by treating imaging data as non-scattering correction components and scattering-only components.

[0016] Figure 3 This is a flowchart depicting another exemplary method for generating radiographic images by treating imaging data as non-scattering correction components and scattering-only components.

[0017] Figure 4 This is a flowchart depicting another exemplary method for generating radiographic images by treating imaging data as non-scattering correction components and scattering-only components.

[0018] Figure 5 This is a flowchart depicting another exemplary method for generating radiographic images by treating imaging data as non-scattering correction components and scattering-only components.

[0019] Figure 6 This is a flowchart depicting another exemplary method for generating radiographic images by treating imaging data as non-scattering correction components and scattering-only components.

[0020] Figure 7A This is an example image generated without scattering correction.

[0021] Figure 7B This is an example image generated using scattering correction.

[0022] Figure 7C This is an example image generated using scattering correction followed by a Gaussian low-pass filter.

[0023] Figure 8A It comes from Figure 7C A larger view of an example image.

[0024] Figure 8B This is an exemplary image generated by processing imaging data as a non-scattering correction component and a scattering-only component.

[0025] Figure 9 It shows the relationship with the source Figures 7A-7C A comparison of exemplary noise measurements associated with an exemplary image of 8B.

[0026] Figure 10 It shows the relationship with the source Figures 7A-7C The exemplary line integral value associated with the exemplary image of 8B along... Figure 9 The comparison of the line profiles shown.

[0027] Figure 11 It shows Figure 10 The diagram shows a partial exploded view of the line integral value.

[0028] Figure 12 A comparison of exemplary images and CT values ​​is shown in relation to conventional image processing and exemplary embodiments that treat imaging data as non-scattering correction components and scattering-only components.

[0029] Figure 13 This is a perspective view of an exemplary imaging apparatus according to one aspect of the disclosed technology.

[0030] Figure 14 This is a schematic diagram of an imaging device integrated into an exemplary radiotherapy apparatus according to one aspect of the disclosed technology.

[0031] Figure 15 This is an exemplary schematic diagram of collimated projection onto a detector.

[0032] Figure 16 This is a flowchart depicting an exemplary method for scattering correction.

[0033] Figure 17 This is a flowchart depicting an exemplary method of IGRT using an imaging device within a radiotherapy apparatus. Detailed Implementation

[0034] The following includes definitions of exemplary terms that may be used throughout this disclosure. Both singular and plural forms of all terms fall within each meaning.

[0035] As used herein, a “component” can be defined as a part of hardware, a part of software, or a combination thereof. A part of hardware may include at least a portion of a processor and memory, where the memory contains instructions to be executed. A component may be associated with a device.

[0036] As used herein, “logic” is synonymous with “circuit” and includes, but is not limited to, hardware, firmware, software, and / or combinations of each performing one or more functions or actions. For example, depending on the desired application or need, logic may include software-controlled microprocessors, discrete logic such as application-specific integrated circuits (ASICs), or other programmable logic devices and / or controllers. Logic may also be entirely software.

[0037] As used herein, "processor" includes, but is not limited to, virtually any number of processor systems or one or more standalone processors, such as any combination of microprocessors, microcontrollers, central processing units (CPUs) and digital signal processors (DSPs), field-programmable gate arrays (FPGAs), and graphics processing units (GPUs). A processor may be associated with a variety of other circuitry that supports its operation, such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), clocks, decoders, memory controllers, or interrupt controllers. This supporting circuitry may be internal or external to the processor or its associated electronic package. The supporting circuitry communicates operatively with the processor. The supporting circuitry is not necessarily shown separately from the processor in block diagrams or other accompanying figures.

[0038] As used herein, “signal” includes, but is not limited to, one or more electrical signals, including analog or digital signals, one or more computer instructions, bits or bit streams, etc.

[0039] As used herein, "software" includes, but is not limited to, one or more computer-readable and / or executable instructions that cause a computer, processor, logic, and / or other electronic device to perform functions, actions, and / or behaviors in a desired manner. Instructions may be embodied in various forms, such as routines, algorithms, modules, or programs that include separate applications or code from dynamically linked sources or libraries.

[0040] Although the exemplary definitions above have been provided, the applicant intends that the broadest reasonable interpretation consistent with this specification be used for these and other terms.

[0041] In CT scans, the X-ray reference data (I0) is the signal when there is no patient (and patient table). When acquiring raw or patient data (I... d When calculating the flux, the ratio of the signal at each detector element is used. If the patient data only has the principal signal (Pr), the logarithm of the ratio is the line integral of the linear decay of the patient along the corresponding X-ray path. The CT image can then be reconstructed from the line integrals measured by all detector elements at many angles around the patient.

[0042] Since the detected signal includes both the primary signal (Pr) and the scattered signal (Sc), where I d =Pr + Sc, the ratio of the reference signal to the detector signal I0 / I d The direct calculation is no longer an integral of the linear decay of the patient along the X-ray path (l), because it is contaminated by the scattered component (Sc) in the signal. Clearly, the correct line integral should be l = log(I0 / Pr). However, with scattering, the calculated ratio is shown in Equation 1:

[0043] ld = log(I0 / I) d ), (1)

[0044] Among them I d =Pr+Sc.

[0045] Since scattering (Sc) is positive and there is no scattering correction, the calculated line integral will be less than the true line integral (without (Sc)). Reconstruction using the contaminated line integral ld will result in quantitative bias in the image, qualitatively reduce contrast, and introduce artifacts into the image.

[0046] To address the aforementioned scattering issues, clinical CT systems can use hardware methods to minimize scattering during initial data acquisition and once data is acquired, as well as software methods to correct residual scattering in the measurement data. The latter can be referred to as scattering correction.

[0047] The principle of scattering correction is to estimate the scattering (Sc_est) and remove or subtract the estimated scattering from the patient data, and then calculate the integral of the correction line according to Equation 2:

[0048]

[0049] If the method is accurate enough that the scattering estimate (Sc_est) is the same as the scattering component in the measurement data, then Equation 2 results in the correct line integral, which will allow for accurate CT image reconstruction.

[0050] However, from the perspective of image noise, scattering correction increases the noise in the calculated line integral, leading to increased noise in the reconstructed image. The variance (noise) of the line integral without scattering correction is shown in Equation 3:

[0051]

[0052] Where Var_Id is the measured patient data I d Let I0 have a variance of 0.

[0053] The variance of the line integral after scattering correction is shown in Equation 4:

[0054]

[0055] Where Var_Sc_est is the variance (noise) of the estimated scattering (Sc_est) (assuming that the noise of the estimated scattering is independent of the noise of the measurement data).

[0056] Comparing the noise of the scattering-corrected line integral in Equation 4 with the noise of the non-scattering-corrected line integral in Equation 3 shows that even if the scattering estimate (Sc_est) is noiseless, the noise of the calculated line integral is amplified by the factor shown in Equation 5:

[0057]

[0058] As the percentage of scattering in the measurement data increases, the noise amplification in Equation 5 increases. For example, if 50% of the measurement data is scattered, then the noise is amplified by a factor of 4. In cone-beam CT systems using an anti-scatter grid, where residual scattering can be 30% of the data, Equation 5 predicts that noise amplification using scatter correction is approximately a factor of 2 compared to no scatter correction. In cone-beam CT with a flat-panel detector, where no anti-scatter grid is deployed, scattering can be greater than 50% of the total measurement data, and the larger the patient, the greater the scattering.

[0059] Scatter correction using conventional noise reduction methods can: (a) reduce noise in the estimated scattering, which corresponds to reducing noise in the second term on the right-hand side of Equation 4; (b) reduce noise in the raw data or line integral after scatter correction; (c) model the scattering in the iterative reconstruction as an additional term to the estimated master image and compare the sum of the estimated master image and the scattering with the measured data; (d) adjust the noise in the reconstruction; or (e) filter / denoise the scatter-corrected image.

[0060] While all these methods may offer some benefits in certain situations, they have numerous drawbacks. For method (a), even if scattering can be perfected without noise, the noise is still amplified by the factor shown in Equation 5, and can be very significant in CT scans with a large amount of scattering, especially for cone-beam CT with a large imaging field of view and no anti-scattering grid. The method in (b) not only suffers from the drawbacks of (a) but may also lose signal (resolution and contrast) because the original data is filtered due to noise reduction. The method in (c) not only requires more complex reconstruction algorithms and longer reconstruction times, but also offers limited noise reduction. The method in (d) faces the challenge of designing the regularization. This challenge can be easily understood if the scatter-corrected image is viewed as a combination of the non-scatter-corrected component and the scatter-corrected component. The non-scatter-corrected component has much lower noise than the scatter-corrected component. Therefore, when applying regularization to the entire image to optimize noise reduction for the high-noise scattering component, it tends to over-regulate the low-noise non-scatter-corrected component of the image. The post-reconstruction image processing (filtering) / denoising in (e) presents the same challenge as in (d) in that it is not possible to optimize two components with very different noise levels.

[0061] In the embodiments disclosed herein, the scattering-corrected image can be processed as a combination of two components: a non-scattering-corrected component and a scattering-only component. The line integral in Equation 2 is rewritten as the sum of the two components, as shown in Equation 6:

[0062]

[0063] The first term on the right-hand side of Equation 6 is the line integral without scattering correction. The term in square brackets is the scattering-corrected component of the line integral. For analytical reconstruction, the reconstruction of the corrected line integral is equivalent to reconstructing the two terms separately to generate two images, and then summing the two images to obtain the final scattering-corrected image. The image reconstructed from the first term is equivalent to a conventional unscattering-corrected image (noSC image). The image reconstructed from the second term can be called a scattering-only image. Specifically, as shown in Equation 7:

[0064] CT image = noSC image + scattering-only image (7)

[0065] In this way, the scattering estimate (Sc_est) can be removed from the patient data. It is clear from the analysis above that scattering-only images carry noise and artifacts associated with scattering correction. NoSC images have significantly lower noise than scattering-only images, and the entire CT image is a combination of both.

[0066] Whether in the original data, the reconstructed image, or during reconstruction, conventional noise reduction associated with scattering correction is essentially performed on both the noSC component and the scattering-only component of the data / image combination, even if these two components have very different noise levels. Methods that optimally suppress noise in the scattering-only component may result in oversmoothing of the noSC component (and thus resolution degradation); and methods that minimize resolution degradation may not effectively suppress the noise associated with scattering correction.

[0067] The embodiments described herein achieve improved image quality after scattering correction, including, for example, sufficient noise reduction and minimal resolution degradation. In these embodiments, processing the scatter-only image separately and differently from the noSC image can reduce noise and artifacts associated with scattering and scattering correction. Due to the much higher noise in the scatter-only image, stronger noise-suppressing data processing techniques (e.g., filters) can be applied to the scatter-only image to optimize noise reduction. However, lighter noise-suppressing data processing techniques (e.g., filters) can be applied to the noSC image to minimize resolution loss. Therefore, by independently optimizing the two image components in Equation 7, the combined final image (e.g., a CT image) can have an optimized trade-off between noise reduction and resolution preservation.

[0068] In some embodiments, using a noSC image to guide noise reduction in a high-noise scatter-only image can have further benefits. Since a noSC image has a much lower noise level than a scatter-only image, it can be used to guide noise reduction in a scatter-only image. For example, "guiding" may include determining the filter kernel and any associated parameters. Such guided noise reduction in a scatter-only image can result in a scatter-only image with similar or even lower noise than the noSC image. Edges in the noSC image can provide reliable edge-preserving guidance for data processing (e.g., filtering) of the scatter-only image. Therefore, the combined final image (e.g., a CT image) can have a noise level similar to that in the noSC image, while edge preservation is optimized.

[0069] In various embodiments, the two components on the right-hand side of Equation 6 are reconstructed differently; for example, a higher resolution filter (kernel) is used to reconstruct the noSC component, and a smoother filter (lower resolution kernel) is used to reconstruct the scattering-only component.

[0070] The following flowcharts and block diagrams illustrate exemplary configurations and methods associated with scattering correction and / or image generation. Exemplary methods can be performed in logic, software, hardware, or combinations thereof. Furthermore, although the processes and methods are presented in a certain order, the blocks can be executed in different orders, including serially and / or in parallel. Thus, the steps including imaging, image-based pre-delivery, and treatment delivery, although shown sequentially, can be performed simultaneously, including in real-time. Additionally, additional steps or fewer steps may be used.

[0071] Figure 1 This is a flowchart depicting an exemplary method 100 for generating a radiographic image by processing imaging data as a non-scattering correction component and a scattering-only component (e.g., processed via data processing techniques, algorithms, filters, etc.). In this way, the data processing techniques applied to the non-scattering correction component and the scattering-only component are separate and distinct, unlike the data processing techniques applied to the combined data. In this embodiment, a patient scan is performed in step 110 to generate radiation / patient data (I...). d 115. A scattering estimate (Sc_est) 117 is also generated. The scattering estimate 117 can be generated in any suitable manner, including based on patient data 115, scattering-only measurements, and / or information associated with the scanning device, including, for example, scan model, parameters, settings, etc. In various embodiments, the scattering-only measurements can come from an effective portion of the radiation detector, for example, by blocking direct radiation (primary data) through a beamformer or collimator.

[0072] In this way, the method then continues to treat the imaging data as a combination of the two components detailed above (as shown in Equation 7): 1) the non-scattering correction component; and 2) the scattering-only component.

[0073] In step 120, the method generates at least one non-scattering correction dataset 145 (e.g., line integral, image, and / or other data) based on patient data 115 and using data processing technique 122 (which may include, for example, filters). In step 130, the method generates at least one scattering-only dataset 155 (e.g., line integral, image, and / or other data) based on scattering estimation 117 and using data processing technique 132 (which may include, for example, another filter). In various embodiments, and as discussed in other embodiments below, one or more types of data processing techniques may be utilized during steps 120 and 130, including during different steps or sub-steps of image data processing.

[0074] As described herein, data processing techniques or steps include software-based mathematical processing of imaging data (e.g., operations applied to data associated with pixels / voxels of image data). Key objectives of applying data processing techniques to imaging data may include noise suppression, preserving spatial resolution and contrast, smoothing, reducing artifacts, and edge enhancement.

[0075] For example, in various embodiments, data processing techniques or steps may include applying one or more filters to the data. In image processing, these filters may include, for example, kernels, convolution matrices, masks, etc. These filters can be used for blurring, sharpening, texturing, edge detection, etc. For example, in several embodiments, this is accomplished by performing convolution between the kernel and the image. Data processing including filtering may be applied to the imaging data before, during, and / or after reconstruction. For example, in one embodiment, radiation / patient data (I d The 115th generation includes raw X-ray data, which consists of the values ​​of all detector signals measured during the CT scan. After calibration, for example, for fluctuations in tube output and beam hardening, the attenuation characteristics of each X-ray signal are considered and correlated with the ray position. Based on this data, the CT image is reconstructed, including mathematical processes such as convolution filtering and backprojection. A convolution filter is a mathematical filtering function (kernel) applied during image reconstruction of the CT imaging data. Reconstruction filters can include sinc filters (e.g., windowing (e.g., Lanczos, Kaiser), splines, etc.), Gaussian, B-splines (e.g., box filters, tent filters), etc.

[0076] Various types of data processing techniques can be selected based on the type of source data (e.g., primary data, primary data and scattering data, scattering-only data, etc.), application (CT, CBCT, PET, SPECT, etc.), desired computational speed, tissue characteristics, etc., including, for example, filters that can be used to smooth or enhance edges. Other types of data processing techniques can include, for example, noise reduction through wavelet transform, singular value decomposition, etc. For example, for singular value decomposition, different eigenvalues ​​can be used for scattering-only and non-scattering components. The reference to filters in the following embodiments is exemplary; other types of data processing techniques can also be used instead of filters or as a supplement to filters.

[0077] In various embodiments, data processing technique 132 differs from data processing technique 122, wherein data processing techniques (e.g., filters) 122, 132 and their associated parameters are specifically targeted at associated data 115, 117. In this way, data processing techniques 122, 132 can be optimized individually for source data 115, 117. For example, utilizing different data processing techniques 122, 132 during the processing of individual data 115, 117 can achieve sufficient noise reduction and minimal resolution degradation. In these embodiments, as described above, processing the scattering estimate 117 (Scatter-only) separately and differently from the patient data 115 (noSC) can improve quality (e.g., reduce noise and artifacts associated with scattering and scattering correction). Specifically, for example, due to the higher noise in the scattering estimate 117, data processing techniques (e.g., filters) with strong noise suppression (e.g., smoothing kernels) 132 can be applied to scattering-only image data to optimize noise reduction. Conversely, data processing techniques (e.g., filters) 122 with lighter noise suppression (e.g., high-resolution kernels) can be applied to the non-scattering-corrected image data to minimize resolution loss. In this way, the two imaging data components (the non-scattering-corrected component and the scattering-only component, as shown in Equation 7) are processed (optimized) independently.

[0078] In some embodiments, the unscattered corrected image / data before or after processing by data processing techniques (e.g., filters) 122 can be used to guide the processing of the relatively high-noise scattering-only image / data by data processing techniques (e.g., filters) 132 (e.g., to determine a filter kernel for noise reduction). As described above, since the unscattered corrected image / data has a much lower noise level than the scattering-only image / data, it can be used to guide noise reduction of the scattering-only image / data in step 130.

[0079] In step 160, the method generates a patient image 165 based on a non-scattering correction dataset 145 and a scattering-only dataset 155 (e.g., by removing scattering from patient data). For example, in an embodiment where the dataset is a line integral component, the scattering-only dataset 155 may be added to the non-scattering correction dataset 145 to produce the patient image 165. In an exemplary embodiment, based on the independent processing at steps 120 and 130, utilizing the corresponding data processing techniques (e.g., filters) 122 and 132 described above, the combined final image 165 (e.g., a CT image) can have an optimized trade-off between noise reduction and resolution preservation.

[0080] Method 100 can be applied to embodiments that process imaging data before or after reconstruction (i.e., in the data or volumetric / image domain), including those detailed in the following embodiments. References to optional filters in the following embodiments are used as exemplary data processing. Other types of data processing techniques may be used without, instead of, or appended to the mentioned exemplary filters.

[0081] Figure 2 This is a flowchart depicting another exemplary method 200 for generating a radiographic image by processing imaging data as a non-scattering correction component and a scattering-only component. In this embodiment, radiation / patient data (I) can be generated as described in method 100. d 115 and scattering estimation (Sc_est) 117, including via patient scan 110.

[0082] In step 221, the method generates at least one nonscattering correction line integral based on patient data 115. Next, in step 222, the nonscattering correction line integral is reconstructed, and in step 225, it is processed to generate a nonscattering corrected image 245. These steps 222, 225 can be performed together or in any order. For example, in this embodiment and other embodiments mentioned below, reconstruction and data processing can be combined (e.g., where data reconstruction can include data processing), can include one or more of each step, and / or can include one or more data processing techniques. In one embodiment, if steps 222, 225 are performed separately (e.g., ...), Figure 2 As shown), each step can utilize the associated filters 223, 226. In another embodiment, if steps 222, 225 are performed together, only filter 223 can be utilized. As described above, in various embodiments, the processing can be performed before, during, and / or after reconstruction. In one embodiment, steps 221, 222, 225 can be associated with an exemplary implementation of step 120.

[0083] In step 231, the method generates at least one scatter-only line integral based on the scattering estimate 117. Next, the scatter-only line integral is reconstructed in step 232 and processed in step 235 to generate a scatter-only image 255. These steps 232 and 235 can be performed together or in any order. In one embodiment, if steps 232 and 235 are performed separately (e.g., ...), ... Figure 2 As shown), each step can utilize associated filters 233, 236. In another embodiment, if steps 232, 235 are performed together, only filter 233 can be utilized. As described above, in various embodiments, the processing can be performed before, during, and / or after reconstruction. In one embodiment, steps 231, 232, 235 can be associated with an exemplary implementation of step 130.

[0084] As described above, in various embodiments, one or more filters 233, 236 differ from one or more filters 223, 226, wherein the filters and their associated parameters may be specifically targeted at the associated data 115, 117. As detailed above, the filters may be optimized for the source data 115, 117 separately, independently optimizing the processing (filtering) of the two imaging data components (non-scattering correction component and scattering-only component, as shown in Equation 7).

[0085] In some embodiments, the nonscattering-corrected image 245 can be used, before or after processing, to guide the processing of relatively high-noise scattering-only images / data, for example, by utilizing filters 233 and / or 236 (i.e., determining filter kernels, for example, for noise reduction). For example, in one embodiment, data processing of the scattering-only data may include applying a Gaussian filter that uses voxel difference in the nonscattering-corrected image 245 to determine kernel weights for the Gaussian filter. When designing filters for the scattering-only data, the nonscattering-corrected image 245 with lower noise can be used to determine (guide) the kernel of a filter (e.g., filter 233 or filter 236) used to filter the scattering-only data to generate a scattering-only image 255. In one embodiment, this can be achieved by either decreasing the kernel weights for pixels on edges (in the nonscattering-corrected image) and increasing the weights, or increasing the kernel size for regions without edges (in the nonscattering-corrected image), and then using the kernel to filter the scattering-only data. As mentioned above, since non-scattering corrected images / data have a much lower noise level than scattering-only images / data, they can be used to guide noise reduction in scattering-only images / data.

[0086] In step 260, the method generates a patient image based on the unscattered corrected image 245 and the scatter-only image 255. For example, according to Equation 7, the scatter-only image 255 can be added to the unscattered corrected image 245 to generate the patient image.

[0087] Figure 3 This is a flowchart depicting another exemplary method 300 for generating a radiographic image by processing imaging data as a non-scattering correction component and a scattering-only component. In this embodiment, radiation / patient data (I) can be generated as described in method 100. d 115 and scattering estimation (Sc_est) 117, including via patient scan 110.

[0088] In step 321, the method generates at least one nonscattering correction line integral based on patient data 115. Next, the nonscattering correction line integral is reconstructed in step 322 and processed in step 325 to generate a nonscattering corrected image 345. These steps 322 and 325 can be performed together or in any order. In one embodiment, if steps 322 and 325 are performed separately (e.g., ...), ... Figure 3 As shown), each step can utilize the associated filters 323, 326. In another embodiment, if steps 322, 325 are performed together, only filter 323 can be utilized. As described above, in various embodiments, the processing can be performed before, during, and / or after reconstruction. In one embodiment, steps 321, 322, 325 can be associated with an exemplary implementation of step 120.

[0089] In step 331, the method generates at least one scattering correction line integral based on patient data 115 and scattering estimation 117. Next, the scattering correction line integral is reconstructed in step 332. Filter 333 can be used before, during, or after reconstruction. Then, in step 334, the method determines the difference between the reconstructed non-scattering correction line integral from step 322 and the reconstructed scattering correction line integral from step 332. For example, in one embodiment, the non-scattering correction image is subtracted from the scattering correction image to produce scattering-only image data. Then, in step 335, the difference can be processed using filter 336 to generate a scattering-only image 355. In one embodiment, steps 331, 332, 334, and 335 can be associated with an exemplary implementation of step 130.

[0090] As described above, in various embodiments, one or more filters 333, 336 differ from one or more filters 323, 326, wherein the filters and their associated parameters may be specifically targeted at the associated data 115, 117. As detailed above, the filters may be optimized for the source data 115, 117 separately, independently optimizing the processing (filtering) of the two imaging data components (non-scattering correction component and scattering-only component, as shown in Equation 7).

[0091] In some embodiments, before or after processing, the non-scattering corrected image 345 can be used to guide the processing of the relatively high-noise scattering-only image / data, for example, by utilizing filter 336 (i.e., determining a filter kernel, for example, for noise reduction). As described above, since the non-scattering corrected image / data has a much lower noise level than the scattering-only image / data, it can be used to guide noise reduction of the scattering-only image in step 335.

[0092] In step 360, the method generates a patient image based on the non-scattering corrected image 345 and the scattering-only image 355. For example, the scattering-only image 355 can be added to the non-scattering corrected image 345 to generate the patient image.

[0093] Figure 4 This is a flowchart depicting another exemplary method 400 for generating a radiographic image by processing imaging data as a non-scattering correction component and a scattering-only component. In this embodiment, radiation / patient data (I) can be generated as described in method 100. d 115 and scattering estimation (Sc_est) 117, including via patient scan 110.

[0094] In step 421, the method generates at least one nonscattering correction line integral based on patient data 115. Next, the nonscattering correction line integral is processed in step 422, and reconstructed in step 425 to generate a nonscattering corrected image 445. These steps 422 and 425 can be performed together or in any order. In one embodiment, if steps 422 and 425 are performed separately (e.g., ...), ... Figure 4 As shown), each step can utilize the associated filters 423, 426. In another embodiment, if steps 422, 425 are performed together, only one filter 426 can be utilized. As described above, in various embodiments, the processing can be performed before, during, and / or after reconstruction. In one embodiment, steps 421, 422, 425 can be associated with an exemplary implementation of step 120.

[0095] In step 431, the method generates at least one scatter-only line integral based on the scattering estimate 117. Next, the scatter-only line integral is processed in step 432, and reconstructed in step 435 to generate a scatter-only image 455. These steps 432 and 435 can be performed together or in any order. In one embodiment, if steps 432 and 435 are performed separately (e.g., ...), ... Figure 4 As shown), each step can utilize the associated filters 433, 436. In another embodiment, if steps 432, 435 are performed together, only filter 436 can be utilized. As described above, in various embodiments, the processing can be performed before, during, and / or after reconstruction. In one embodiment, steps 431, 432, 435 can be associated with an exemplary implementation of step 130.

[0096] As described above, in various embodiments, one or more filters 433, 436 differ from one or more filters 423, 426, wherein the filters and their associated parameters may be specifically targeted at the associated data 115, 117. As detailed above, the filters may be optimized for the source data 115, 117 separately, independently optimizing the processing (filtering) of the two imaging data components (non-scattering correction component and scattering-only component, as shown in Equation 7).

[0097] In some embodiments, the nonscattering correction line integral from step 421, before or after processing, can be used to guide the processing of the relatively high-noise scattering-only data, for example, by utilizing filters 433 and / or 436 (e.g., to determine a filter kernel for noise reduction). As described above, since the nonscattering correction data has a much lower noise level than the scattering-only data, it can be used to guide noise reduction of the scattering-only data.

[0098] In step 460, the method generates a patient image based on the non-scattering corrected image 445 and the scattering-only image 455. For example, the scattering-only image 455 can be added to the non-scattering corrected image 445 to generate the patient image.

[0099] Figure 5 This is a flowchart depicting another exemplary method 500 for generating a radiographic image by processing imaging data as a non-scattering correction component and a scattering-only component. In this embodiment, radiation / patient data (I) can be generated as described in method 100. d 115 and scattering estimation (Sc_est) 117, including via patient scan 110.

[0100] In step 521, the method generates at least one nonscattering correction line integral based on patient data 115. Next, the nonscattering correction line integral is processed in step 522, and reconstructed in step 525 to generate a nonscattering corrected image 545. These steps 522, 525 can be performed together or in any order. In one embodiment, if steps 522, 525 are performed separately (e.g., ...), ... Figure 5 As shown), each step may utilize associated filters 523, 526. In another embodiment, if steps 522, 525 are performed together, only one filter 526 may be utilized. As described above, in various embodiments, the processing may be performed before, during, and / or after reconstruction. In one embodiment, steps 521, 522, 525 may be associated with an exemplary implementation of step 120.

[0101] In step 531, the method generates at least one scattering correction line integral based on patient data 115 and scattering estimation 117. Next, the scattering correction line integral is reconstructed in step 532. Filter 533 can be used before, during, or after reconstruction. Then, in step 534, the method determines the difference between the reconstructed non-scattering correction line integral from step 525 and the reconstructed scattering correction line integral from step 532. For example, in one embodiment, the non-scattering correction image is subtracted from the scattering correction image to produce scattering-only image data. Then, in step 535, the difference can be processed using filter 536 to generate a scattering-only image 555. In one embodiment, steps 531, 532, 534, and 535 can be associated with an exemplary implementation of step 130.

[0102] As described above, in various embodiments, one or more filters 533, 536 differ from one or more filters 523, 526, wherein the filters and their associated parameters may be specifically targeted at the associated data 115, 117. As detailed above, the filters may be optimized for the source data 115, 117 separately, independently optimizing the processing (filtering) of the two imaging data components (non-scattering correction component and scattering-only component, as shown in Equation 7).

[0103] In some embodiments, the non-scattering corrected image 545 can be used before or after processing to guide the processing of the relatively high-noise scattering-only image / data, for example, by utilizing filter 536 (i.e., determining the filter kernel, for example, for noise reduction). As described above, since the non-scattering corrected image / data has a much lower noise level than the scattering-only image / data, it can be used to guide noise reduction of the scattering-only image in step 535.

[0104] In step 560, the method generates a patient image based on the unscattered corrected image 545 and the scatter-only image 555. For example, the scatter-only image 555 may be added to the unscattered corrected image 545 to generate the patient image.

[0105] Figure 6 This is a flowchart depicting another exemplary method 600 for generating a radiographic image by processing imaging data as a non-scattering correction component and a scattering-only component. In this embodiment, radiation / patient data (I) can be generated as described in method 100. d 115 and scattering estimation (Sc_est) 117, including via patient scan 110.

[0106] In step 621, the method generates at least one non-scattering correction line integral based on patient data 115. Next, in step 622, the non-scattering correction line integral is processed using filter 623. In step 631, the method generates at least one scattering-only line integral based on scattering estimation 117. Next, in step 632, the scattering-only line integral is processed using filter 633.

[0107] As described above, in various embodiments, one or more filters 633 differ from one or more filters 623, wherein the filters and their associated parameters may be specifically targeted at the associated data 115, 117. As detailed above, the filters may be optimized for the source data 115, 117 separately, independently optimizing the processing (filtering) of the two imaging data components (non-scattering correction component and scattering-only component, as shown in Equation 7).

[0108] In some embodiments, the nonscattering correction line integral from step 621 can be used, before or after processing, to guide the processing of the relatively high-noise scattering-only data, for example, by utilizing filter 633 (i.e., determining a filter kernel, for example, for noise reduction). As described above, since the nonscattering correction data has a much lower noise level than the scattering-only data, it can be used to guide noise reduction of the scattering-only data.

[0109] Then, in step 640, the method separates the principal data, for example, by determining the difference between the non-scattering correction line integral from steps 621 and 622 and the scattering-only line integral from steps 631 and 632. For example, in one embodiment, the scattering-only line integral data is added to the non-scattering correction line integral data to produce only principal line integral data. Then, in step 650, the principal line integral data can be reconstructed, including using filter 655. In step 660, the method generates patient image 665.

[0110] Figure 7-11 illustrates the performance of an exemplary cone-beam CT scan of a pelvic phantom. Compared to conventional methods utilizing post-reconstruction low-pass filters, the results of this example demonstrate significantly improved preservation of bone boundaries (edges in the image) and similar noise reduction. Moreover, the noise patterns are much more natural than in low-pass filtered images.

[0111] In particular, Figures 7A-7C 700 cone-beam CT images of a pelvic model acquired using a collimator aperture of approximately 10 cm, at the isocenter, at 125 kVp, 2.5 mAs, with 360-degree views / rotations and 24 views per second. Figures 7A-7CThe image quality achieved using conventional image processing without utilizing the aforementioned two-component image processing techniques (i.e., using personalized filters to process the non-scattering correction component and the scattering-only component separately, as shown in the above method) is demonstrated.

[0112] Figure 7A Image 710 is without scattering correction. Figure 7B It is an image 720 that uses scattering correction. Figure 7C Image 730 is scatter-corrected using a Gaussian low-pass filter. As shown in these images, the scatter-correction employed in image 720 minimizes deep-pull artifacts 712 (i.e., the decrease in intensity in the center of the phantom compared to the periphery in image 710), but image noise is amplified and fringe artifacts are introduced throughout image 720. Post-reconstruction processing using a low-pass filter in image 730 reduces the magnitude of noise and fringe artifacts, but at the expense of image resolution (e.g., noticeably blurred bone boundaries).

[0113] Figures 8A-8B A comparison 800 of image quality achieved using conventional image processing and implementing the aforementioned two-component image processing technique (i.e., processing the non-scattering correction component and the scattering-only component separately with personalized filters) is shown. Specifically, Figure 8A It comes from Figure 7C The larger view of image 730 uses conventional scattering correction followed by a Gaussian low-pass filter. Figure 8B Image 810 of the same pelvic phantom using the method 200 described above shows reduced noise and artifacts and improved edge preservation performance compared to the Gaussian filtered image 730.

[0114] Specifically, in order to generate image 810, a noSC-based image (e.g., Figure 2 In method 245), the scatter-only image is processed (filtered), where the noSC image is used as a guide image to determine or compute the filter kernel. As in method 200, the scatter-only image is processed (filtered) using a different filter than the one used for the noSC image. In this embodiment, the filter kernel is a local Gaussian kernel of size 7×7×7, and the weight of each voxel is computed using the HU difference between that voxel and the central voxel in the noSC image. Edge information from the noSC image is naturally incorporated into the filter (kernel) computation for the scatter-only image to preserve corresponding edges in the scatter-only image. The resulting image 810 demonstrates effective edge preservation as well as noise and artifact reduction.

[0115] Figure 9A comparison 900 of noise measurements associated with images 710, 720, 730, and 810 is shown. Specifically, the average level and noise measurements for different regions of interest (ROIs) from images 710, 720, 730, and 810 from different processing are included in Table 910. Note that the average values ​​in Table 910 are CT numbers plus 1000. The noSC noise data, SC noise data, and SC+Gaussian (SC+Gau) noise data are respectively compared with… Figures 7A-7C The three images shown, 710, 720, and 730, are associated. The "example" noise data is related to... Figure 8B The image 810 shown is associated with this. The regions of interest include the ROI_middle 912, the ROI_outer periphery 914, and the line contour 916 (in...). Figure 10 (See detailed diagram). The noise level of image 810 is similar to that of noSC image 710 in ROI_outer periphery 914, and lower than that of noSC image 710 in ROI_middle 912. SC+Gau image 730 has similar noise to image 810 in ROI_middle 912, but lower noise in ROI_outer periphery 914. However, as shown in Figure 8, SC+Gau image 730 has very strong residual fringe artifacts, which are minimal in image 810. Line contour 916 is used for... Figure 10 The line profiles shown are compared.

[0116] Figure 10 The comparison of line integral values ​​associated with images 710, 720, 730, and 810 along line contour 916 is shown in 1000. Note that the values ​​in the line integral are the CT number plus 1000. The “NoSc” line, “Sc” line, and “Sc_GaussianPF (Sc_GaussianPF)” line are respectively compared with those passing through... Figures 7A-7C The line outline 916 of the three images 710, 720, and 730 shown is associated with the example line. Figure 8B The line outline 916 shown is associated with the image 810. Figure 11 A decomposed plot 1100 of portion 1010 of the data is shown. The line profile also demonstrates the effectiveness of the exemplary embodiment in noise and fringe artifact suppression and edge preservation. For example, the Sc line has exaggerated peaks and valleys when compared to the noSc line. Moreover, the Sc_GaussianPF line is too smooth in these regions. In contrast, the example line has removed scattering components, but without the associated noise, artifacts, resolution loss, etc.

[0117] Figure 12A comparison 1200 of images and CT values ​​associated with conventional image processing and exemplary embodiments of the aforementioned two-component image processing technique in a Caphan scan with an annulus is shown. This comparison 1200 demonstrates significant noise reduction while preserving small object boundaries (visual assessment). Image 1210 (non-SC) has no scattering correction. Image 1220 (SC) has scattering correction. Image 1230 (example) is scattering corrected according to the aforementioned two-component method. To generate the images, a CatPhan scan with an annulus is performed at isogonal points using an aperture of 4.5 cm. Images 1210, 1220, and 1230 are displayed in a HU window [-400, 200]. Scan parameters are 125 kV, 2.5 mAs, 480-degree view / rotation, and 24 frames / second. Table 1250 shows the average CT values ​​plus / minus standard deviation for ROI 1212 (center—displayed as a circle), ROI 1214 (middle—displayed as a dashed ellipse), and ROI 1216 (outer periphery—displayed as a solid ellipse), where the average is the CT number plus 1000. Image 1230 (example) using the exemplary embodiment has more noise suppression compared to the scatter-corrected image 1220 (SC). In both images 1220 and 1230, small contrast boundaries and contrast are visually identical.

[0118] As discussed in detail above, embodiments of the disclosed technology relate to correcting scattering in imaging data, including utilizing patient data from imaging scans (I d ) and scattering estimation (Sc_est). Imaging scans can be performed by any radiographic imaging device associated with the scan type, including X-ray, CT, CBCT, SPECT, PET, MR, etc. These methods can be used for scattering correction in imaging data from these imaging scans, for example, for noise and artifact reduction. Although CT scanners and cone-beam CT scanners are emphasized in several exemplary embodiments, this technique can also be applied to image reconstruction / data processing based on removing unwanted counts / signals from the initial counts to generate corrected images, such as scattering correction in SPECT, PET, MR, SPECT / CT, PET / MR, etc.

[0119] In various embodiments, imaging scans can be performed using a dedicated imaging device or an imaging device integrated with a radiotherapy delivery device. For example, a radiotherapy delivery device can utilize an integrated low-energy radiation source for CT to be used in conjunction with or as part of IGRT. In particular, for example, radiotherapy delivery devices and associated methods can combine a low-energy collimated radiation source for imaging in a gantry using rotational (e.g., spiral or step-and-shoot) image acquisition with a high-energy radiation source for therapeutic treatment, as described in U.S. Patent Application Serial No. 16 / 694,145, filed November 25, 2019, entitled “Multimodal Radiation Apparatus and Method,” and U.S. Patent Application Serial No. 16 / 694,148, filed November 25, 2019, entitled “Apparatus and Method for Scalable Field Imaging Using a Multi-Source System,” the entire contents of which are incorporated herein by reference. In these embodiments, low-energy radiation sources (e.g., kilovolts (kV)) can produce higher quality images than imaging using high-energy radiation sources (e.g., megavolts (MV)).

[0120] Imaging data acquisition methods may include or otherwise utilize multi-rotation scanning, which may be, for example, continuous scanning (e.g., having a helical source trajectory around a central axis and longitudinal movement of the patient support through a gantry hole), discontinuous circular stop-and-reverse scanning with incremental longitudinal movement of the patient support, step-and-shoot circular scanning, etc.

[0121] According to various embodiments, the imaging apparatus uses, for example, a beamformer to collimate a radiation source into, for example, a cone beam or a fan beam. In one embodiment, the collimated beam may be combined with a gantry that rotates continuously as the patient moves, resulting in helical image acquisition.

[0122] In various embodiments, detectors (with various row / slice sizes, configurations, dynamic ranges, etc.), scan spacing, and / or dynamic collimation are additional features, including selectively exposing portions of the detector and selectively defining effective readout areas, as discussed in detail below.

[0123] Imaging apparatuses and methods can provide selective and variable collimation of a radiation beam emitted by a radiation source, including adjusting the shape of the radiation beam to expose a smaller-than-usual effective area than that of the associated radiation detector (e.g., a radiation detector positioned to receive radiation from the radiation source). Exposing only the primary zone of the detector to direct radiation allows the shadowed zone of the detector to receive only scattered radiation. Measurements of scattering in the shadowed zone of the detector (and in some embodiments, in the penumbra) can be used to estimate the scattering in the primary zone of the detector receiving projection data.

[0124] refer to Figure 13 and 14 An exemplary imaging device 10 is shown (which may include, for example, an X-ray imaging device). It should be understood that the imaging device 10 can be used with radiotherapy equipment (such as...) Figure 14 The imaging apparatus 10 (shown) is associated with and / or integrated into a radiotherapy device that can be used for various applications, including but not limited to IGRT. The imaging apparatus 10 includes a rotatable gantry system, referred to as gantry 12, supported by a support unit or housing 14 or otherwise housed therein. As used herein, gantry refers to a gantry system comprising one or more gantry structures (e.g., rings or C-arms) capable of supporting one or more radiation sources and / or associated detectors as the gantry rotates around a target. The rotatable gantry 12 defines a gantry aperture 16 through which a patient can move into and be positioned for imaging and / or treatment. According to one embodiment, the rotatable gantry 12 is configured as a slip-ring gantry to provide continuous rotation of the imaging radiation source (X-ray) and associated radiation detector while providing sufficient bandwidth for high-quality imaging data received by the detector.

[0125] The patient support 18 is positioned adjacent to the rotatable stage 12 and configured to support the patient in a generally horizontal position for longitudinal movement into and within the rotatable stage 12. The patient support 18 can move the patient, for example, in a direction perpendicular to the plane of rotation of the stage 12 (along or parallel to the axis of rotation of the stage 12). The patient support 18 can be operatively coupled to a patient support controller for controlling the movement of the patient and the patient support 18. The device 10 is capable of performing volume-based and planar imaging acquisition. For example, in various embodiments, the device 10 can be used to acquire volumetric images and / or planar images and perform the associated processing methods described above.

[0126] like Figure 14 As shown, the imaging apparatus 10 includes an imaging radiation source 30 coupled to or otherwise supported by a rotatable stage 12. The imaging radiation source 30 emits a radiation beam (generally designated 32) for generating high-quality images. In this embodiment, the imaging radiation source is an x-ray source 30 configured as a kilovolt (kV) source (e.g., a clinical x-ray source with energy levels in the range of approximately 20 kV to approximately 150 kV). The imaging radiation source can be any type of transmission source suitable for imaging. In various other embodiments, other imaging transmission sources can be used interchangeably.

[0127] The imaging apparatus 10 may also include another radiation source 20 coupled to or otherwise supported by the rotatable stage 12. According to one embodiment, the radiation source 20 is configured as a therapeutic radiation source, such as a high-energy radiation source for treating tumors within a patient in a region of interest. It should be understood that the therapeutic radiation source may be a high-energy X-ray beam (e.g., a megavolt (MV) X-ray beam). Typically, the radiation source 20 has a higher energy level (peak and / or average, etc.) than the imaging radiation source 30. Although... Figure 13 and 14 An X-ray imaging apparatus 10 with a radiation source 30 mounted to a ring gantry 12 is depicted, but other embodiments may include other types of rotatable imaging apparatus, including, for example, C-arm gantry and robotic arm-based systems.

[0128] Detector 34 (e.g., a two-dimensional planar detector or a curved surface detector) may be coupled to or otherwise supported by a rotatable stage 12. Detector 34 (e.g., an X-ray detector) is positioned to receive radiation from imaging radiation source 30 and may rotate with source 30. Detector 34 may detect or otherwise measure the amount of unattenuated radiation, thus inferring the radiation actually attenuated by the patient or associated patient ROI (by comparison with the initially generated phase). As radiation source 30 rotates and emits radiation toward the patient, detector 34 may detect or otherwise collect attenuation data from different angles.

[0129] A collimator or beamformer assembly (generally designated 36) is positioned relative to the imaging source 30 to selectively control and adjust the shape of the radiation beam 32 emitted by the source 30 to selectively expose a portion or region of the effective area of ​​the detector 34. The beamformer can also control how the radiation beam 32 is positioned on the detector 34. For example, in one embodiment, 3-4 cm of projected image data can be captured with each readout, with an unexposed detector area of ​​approximately 1-2 cm on one side or each side, which can be used to capture scattered data.

[0130] Detector 24 may be coupled to or otherwise supported by a rotatable stage 12 and is positioned to receive radiation 22 from the therapeutic radiation source 20. Detector 24 may detect or otherwise measure the amount of unattenuated radiation, thus inferring the radiation actually attenuated by the patient or associated patient ROI (by comparison with the initially generated radiation). As the therapeutic radiation source 20 rotates and emits radiation toward the patient, detector 24 may detect or otherwise collect attenuation data from different angles.

[0131] The therapeutic radiation source 20 can be installed, configured, and / or moved to the same plane as or a different plane (offset) from the imaging source 30. In some embodiments, offsetting the radiation plane can reduce scattering caused by the simultaneous activation of the radiation sources 20 and 30.

[0132] When integrated with a radiotherapy device, the imaging apparatus 10 can provide images for setting up (e.g., aligning and / or registering), planning, and / or guiding the radiation delivery process (treatment). Typical setup is accomplished by comparing current (during treatment) images with pre-treatment image information. Pre-treatment image information may include, for example, X-ray, CT, CBCT, MR, PET, SPECT, and / or 3D rotational angiography (3DRA) data, and / or any information obtained from these or other imaging modalities. In some embodiments, the imaging apparatus 10 can track patient, target, or ROI movement during treatment.

[0133] The reconstruction processor 40 may be operatively coupled to detectors 24, 34. In one embodiment, the reconstruction processor 40 is configured to generate patient images based on radiation received from radiation sources 20, 30 by detectors 24, 34, as described above. It will be understood that the reconstruction processor 40 may be configured to perform the methods described herein. The apparatus 10 may also include a memory 44 suitable for storing information including, but not limited to, data processing and reconstruction algorithms and software, including filters and data processing / filter parameters, imaging parameters, image data from previously or otherwise previously acquired images (e.g., planning images), treatment plans, etc.

[0134] Imaging apparatus 10 may include an operator / user interface 48, through which an operator of imaging apparatus 10 can interact with or otherwise control imaging apparatus 10 to provide input related to scan or imaging parameters. Operator interface 48 may include any suitable input device, such as a keyboard, mouse, voice-activated controller, etc. Imaging apparatus 10 may also include a display 52 or other human-readable element to provide output to the operator of imaging apparatus 10. For example, display 52 may allow the operator to view reconstructed patient images and other information related to the operation of imaging apparatus 10, such as imaging or scan parameters.

[0135] like Figure 14As shown, the imaging apparatus 10 includes a controller (generally designated 60) operatively coupled to one or more components of the apparatus 10. The controller 60 controls the overall function and operation of the apparatus 10, including providing power and timing signals to the imaging source 30 and / or the therapeutic radiation source 20, and a gantry motor controller controlling the rotational speed and position of the rotatable gantry 12. It will be understood that the controller 60 may include one or more of the following: a patient support controller, a gantry controller, a controller coupled to the therapeutic radiation source 20 and / or the imaging source 30, a beamformer 36 controller, a controller coupled to the detector 24 and / or the detector 34, etc. In one embodiment, the controller 60 is a system controller capable of controlling other components, devices, and / or controllers.

[0136] In various embodiments, the rebuild processor 40, operator interface 48, display 52, controller 60 and / or other components may be combined into one or more components or devices.

[0137] Device 10 may include various components, logic, and software. In one embodiment, controller 60 includes a processor, memory, and software. By way of example and not limitation, imaging apparatus and / or radiotherapy systems may include a variety of other devices and components (e.g., gantry, radiation source, collimator, detector, controller, power supply, patient support, etc.) that can implement one or more routines or steps associated with imaging and / or IGRT for a particular application, wherein routines may include imaging, image-based pre-delivery steps, and / or treatment delivery, including corresponding device settings, configurations, and / or locations (e.g., paths / trajectories) that may be stored in memory. Furthermore, controller(s) may directly or indirectly control one or more devices and / or components based on one or more routines or procedures stored in memory. Examples of direct control include setting various radiation source or collimator parameters (power, velocity, position, timing, modulation, etc.) associated with imaging or treatment. Examples of indirect control include transmitting position, path, velocity, etc., to a patient support controller or other peripheral device. The hierarchy of various controllers that may be associated with the imaging apparatus may be arranged in any suitable manner to transmit appropriate commands and / or information to the desired devices and components.

[0138] Furthermore, those skilled in the art will understand that other computer system configurations can be used to implement the system and method. The aspects of the invention illustrated can be implemented in a distributed computing environment, where certain tasks are performed by local or remote processing devices linked via a communication network. For example, in one embodiment, the reconstruction processor 40 may be associated with a separate system. In a distributed computing environment, program modules may reside in both local and remote memory storage devices. For example, a remote database, a local database, a cloud computing platform, a cloud database, or a combination thereof may be utilized in conjunction with the imaging device 10.

[0139] Imaging apparatus 10 may utilize exemplary environments for implementing various aspects of the invention, including a computer, wherein the computer includes a controller 60 (e.g., including a processor and memory, which may be memory 44) and a system bus. The system bus may couple system components, including but not limited to memory, to the processor and may communicate with other systems, controllers, components, devices, and the processor. Memory may include read-only memory (ROM), random access memory (RAM), hard disk drives, flash drives, and any other form of computer-readable medium. Memory may store various software and data, including routines and parameters, which may include, for example, treatment plans.

[0140] Many factors determine image quality (e.g., image source focal spot size, detector dynamic range, etc.). Scattering is a limiting factor in many imaging techniques and image quality. Various methods can be used to reduce scattering. One approach is to use an anti-scattering grid (which collimates the scattering). However, implementing a scattering grid on a kV imaging system can be problematic, including for motion tracking and correction. As mentioned above, accurately estimating the scattering in the projection data is essential for improving the quality of the image data. In various embodiments, the scattering in the projection data acquired in the main partitions of detector 34 can be estimated based on data measured in the shadow partitions (and penumbra partitions) of detector 34.

[0141] Figure 15 This is a schematic diagram of an exemplary collimated projection 1500 onto detector 1502. A rotating radiation source 1506 (e.g., X-rays) is shown emitting a radiation beam 1508 exposed to a primary or central (C) section 1510 of detector 1502 to guide radiation from source 1506 (e.g., through a target) as source 1506 rotates about the y-axis. Patient support (not shown) movement can be in an axial (longitudinal) direction along the y-axis, including as part of a scan as described above. Detector 1502 also has a rear (B) shadow section 1512 and a front (F) shadow section 1514, which are blocked by a beamformer / collimator 1520 from direct exposure to the radiation beam 1508. The beamformer / collimator 1520 is configured to adjust the shape and / or position of the radiation beam 1508 emitted from source 1506 onto detector 1502. Shadow sections 1512, 1514 will receive only scattered radiation.

[0142] The collimator 1520 opening is configured such that the rear (B) end 1512 and front (F) end 1514 of the detector 1502 in the axial or longitudinal direction (along the patient table direction or the y-axis) are not directly illuminated by radiation 1508. These rear (B) 1512 (in the negative longitudinal direction along the rotational y-axis) and front (F) 1514 (in the positive longitudinal direction along the rotational y-axis) can be utilized for scattering measurements because they do not receive direct radiation. For example, the readout range of the detector 1502 can be configured to read all or part of the data in one or more shaded zones 1512, 1514, and use this data for scattering estimation in the main zone 1510. The main or central (C) zone 1510 receives both direct projection and scattering.

[0143] In various embodiments, a data processing system (including, for example, processor 40) may be configured to receive projection data measured in a primary partition 1510 and scattering data measured in at least one shadow partition 1512, 1514, and then determine an estimated scattering in the primary partition 1510 based on the scattering data measured in at least one shadow partition 1512, 1514. In some embodiments, determining the estimated scattering in the primary partition 1510 during the current rotation may be based on scattering data measured in at least one shadow partition 1512, 1514 during adjacent (previous and / or subsequent) rotations. In other embodiments, measurement data from one or more penumbra partitions (adjacent to the primary and shadow partitions) may also be used for scattering estimation.

[0144] Various techniques and methods can utilize different scanning geometries, detector positioning, and / or beamformer window shapes. In some embodiments, the detector may also be offset in the lateral direction.

[0145] Figure 16 This is a flowchart depicting an exemplary method 1600 for scattering estimation and correction, as described above. Inputs may include any optional prior data and / or scan design. In this embodiment, step 1610 includes data acquisition. For example, during the rotation of a radiation source projecting a collimated radiation beam toward a target and a radiation detector, the method measures projection data (primary + scattering) in the central (primary) partition of the radiation detector and uses the front and / or rear shadow periphery partitions of the detector to measure scattering. Data acquisition in step 1610 may also include adjusting the shape / position of the radiation beam and / or adjusting the readout range (including determining the effective partition) using a beamformer before and / or during the scan.

[0146] Next, step 1620 includes scattering estimation. For example, the method uses scattering measurements from one or more shadowed zones to estimate scattering in the projection data from the central (primary) zone. Then, step 1630 includes scattering correction, which may include any of the two-component techniques described above. The output includes scatter-corrected projection data suitable for imaging. Various embodiments may utilize different scanning geometries, detector positioning / effective partitioning, beamformer positioning / window shape, etc.

[0147] Figure 17 This is a flowchart depicting an exemplary method 1700 of IGRT using a radiotherapy apparatus (including, for example, imaging device 10). Previous image data 1705 of the patient may be available, which may be previously acquired planning images, including previous CT images. Previous data 1705 may also include treatment plans, phantom information, models, prior information, etc. In some embodiments, previous image data 1705 is generated by the same radiotherapy apparatus, but at an earlier time. In step 1710, imaging of the patient is performed using a low-energy radiation source (e.g., kV radiation from x-ray source 30). In one embodiment, imaging includes a helical scan with a fan-beam or cone-beam geometry. Step 1710 may use the scattering estimation and correction techniques described above to generate high-quality (HQ) images or imaging data 1715. In some embodiments, image quality may be adjusted to optimize the balance between image quality / resolution and dose. In other words, not all images need to have the highest quality, or image quality may be adjusted to optimize or compromise the balance between image quality / resolution and image acquisition time. Imaging step 1710 may also include image / data processing to generate a patient image based on the imaging data (e.g., according to the methods described above). Although image processing step 1720 is shown as part of imaging step 1710, in some embodiments, image processing step 1720 is a separate step, including where image processing is performed by a separate device.

[0148] Next, in step 1730, one or more image-based pre-delivery steps, discussed below, are performed, at least in part, based on the imaging data 1715 from step 1710. As discussed in more detail below, step 1730 may include determining various parameters associated with therapeutic treatment and (subsequent) imaging planning. In some embodiments, the image-based pre-delivery step (1730) may require further imaging (1710) prior to treatment delivery (1740). Step 1730 may include adjusting the treatment plan based on the imaging data 1715 as part of an adaptive radiotherapy routine. In some embodiments, the image-based pre-delivery step 1730 may include real-time treatment planning. Embodiments may also include simultaneous, overlapping, and / or alternating activation of imaging and therapeutic radiation sources. Real-time treatment planning may involve any or all of these types of imaging and therapeutic radiation activation techniques (simultaneous, overlapping, and / or alternating).

[0149] Next, in step 1740, therapeutic treatment delivery is performed using a high-energy radiation source (e.g., MV radiation from therapeutic radiation source 20). Step 1740 delivers a therapeutic dose 1745 to the patient according to the treatment plan. In some embodiments, the IGRT method 1700 may include returning to step 1710 to perform additional imaging at various intervals, followed by an image-based pre-delivery step (1730) and / or treatment delivery (1740) as needed. In this way, a device 10 capable of adaptive treatment can be used to generate and utilize high-quality imaging data 1715 during IGRT. As described above, steps 1710, 1720, 1730, and / or 1740 may be performed simultaneously, overlapping, and / or alternately.

[0150] In various embodiments, the above methods can be used for scatter correction, regardless of whether the imaging data is generated using a dedicated imaging device or an imaging device integrated with a radiotherapy delivery device.

[0151] In one embodiment, the CT apparatus includes a rotating X-ray source and an X-ray detector, which acquire raw data (e.g., I0) for generating CT images. d A set of hardware and / or software is used to measure and / or generate a set of scattering data (e.g., Sc_est) to compensate for / correct scattering contamination in the original data. An unscattered corrected image is reconstructed from the original data, and a scattering-only image is reconstructed from the scattering data. In this embodiment, the original data can be used to calculate the unscattered corrected line integral to reconstruct the unscattered corrected CT image. The scattering data can be used to calculate the scattering-only line integral based on Equation 6 to reconstruct the scattering-only image. The unscattered corrected image and the scattering-only image are processed independently, with the latter being more heavily filtered due to higher noise. The processed unscattered corrected image and the processed scattering-only image can be combined to create a final CT image with scattering correction.

[0152] In another embodiment, volumetric image subtraction can be used to generate a scatter-only image. Here, scattering data is used with the original data to generate a scattering correction line integral for reconstructing the scattering-corrected image. The original data can be used to compute the non-scattering correction line integral to reconstruct the non-scattering-corrected image. The non-scattering-corrected image can be subtracted from the scattering-corrected image to obtain the scattering-only image. The non-scattering-corrected image and the scattering-only image are processed independently, with the latter being more heavily filtered due to higher noise. The processed non-scattering-corrected image and the processed scattering-only image can be combined to create a final CT image with scattering correction.

[0153] In various embodiments, the nonscattering-corrected image can be used to guide the processing of the scattering-only image to achieve effective noise and artifact reduction while preserving edges in the image. For example, the filter could be a Gaussian filter that uses voxel differences in the nonscattering-corrected image to determine the kernel weights of the scattering-only image filter. In this way, edge information in the nonscattering-corrected image is used to preserve corresponding edges in the scattering-only image. The nonscattering-corrected image can also be used in more advanced edge-preserving processing schemes to enhance the processing of the scattering-only image. For example, processing the scattering-only image can be based on anisotropic differential filter parameters obtained in the nonscattering-corrected image.

[0154] In another embodiment, different reconstruction schemes can be used to reconstruct the unscattered corrected image and the scattering-only image. For example, a higher resolution kernel than that used for the scattering-only image can be used to reconstruct the unscattered corrected image, and a custom-designed fringe artifact reduction algorithm can be used to reconstruct the scattering-only image. Different raster arrays can be used to reconstruct the scattering-only image to accelerate reconstruction time. For example, if a 512×512 matrix is ​​used for unscattered corrected image reconstruction, a 256×256 matrix can be used for scattering-only image reconstruction to accelerate reconstruction time. The reconstructed scattering-only image can then be resampled to the same raster array as the unscattered corrected image. The unscattered corrected image can then be used to guide the processing of the scattering-only image. The resulting scattering-only image can be combined with the unscattered corrected image to create a final image with scattering correction.

[0155] In addition to the CT environment emphasized in several exemplary embodiments, in various other embodiments, the acquisition or generation of images with scattering (e.g., IC) is performed. dVarious imaging devices can use scattering data (e.g., Sc_est) to correct raw data from SPECT, PET, etc. Scattering data can be used to modify / correct line integrals, where the line integral can be decomposed into components without scattering correction and linear combinations of components with scattering correction, similar to Equation 6. Two images can be reconstructed separately to optimize their quality, and then combined to obtain the final image. The reconstructed unscattered corrected image and the scattering-only image can be processed independently to optimize their quality, and then combined to obtain the final image. The unscattered corrected image can also be used as a guide image to determine the weights of the filter kernel when processing the scattering-only image.

[0156] Besides embodiments that utilize the unscattered corrected image to guide the processing of the scatter-only image (i.e., operate in the image domain), other embodiments can operate in the data domain. In these embodiments, processing of the line integral of the generated scatter-only component can be based on the line integral data of the unscattered corrected component as guiding data to preserve edges in the scatter-only component. The resulting line integral of the scatter-only component can be reconstructed separately or together with the line integral of the unscattered corrected component.

[0157] In various embodiments, the original data (e.g., I) d The scattering data, along with measured scattering data (e.g., Sc_est), is used to reconstruct the scatter-corrected image, and the raw data is used to reconstruct the unscatter-corrected image using various reconstruction algorithms to obtain the final image. In some embodiments, the reconstruction may be analytical reconstruction. In some embodiments, the reconstruction may be iterative reconstruction. In various embodiments, the scattering-only image is processed separately from the unscatter-corrected image (including filtering, artifact reduction, etc.) and then combined with the unscatter-corrected image to obtain the final image. In some embodiments, the scattering-only image is generated by subtracting the unscatter-corrected image from the scatter-corrected image. Furthermore, the unscatter-corrected image can be used to guide the processing of the scattering-only image to achieve optimal noise and artifact reduction and edge preservation.

[0158] Typically, in various embodiments, the above-described techniques can be applied to any imaging device, and any correction method used for the line integral for image reconstruction generates correction terms that alter the line integral used for image reconstruction (e.g., leading to increased image noise and artifacts). For example, the correction term could be a hysteresis correction term in cone-beam CT using a flat-panel detector. Multiple correction terms, such as hysteresis correction and scattering correction in cone-beam CT, collectively alter the line integral used for reconstruction, while the line integral can be decomposed into two components, uncorrected and corrected, similar to the correction in Equation 6. The above-described methods can be utilized to obtain a final image with improved quality and performance.

[0159] Although the disclosed technology has been shown and described with respect to specific aspects, embodiments, or multiple embodiments, it will be apparent to those skilled in the art, upon reading and understanding this specification and the accompanying drawings, that equivalent changes and modifications will occur. In particular, with respect to the various functions performed by the foregoing elements (parts, components, devices, elements, formations, etc.), unless otherwise indicated, the terminology used to describe these elements (including references to “means”) is intended to correspond to any element performing the specified function of the described element (i.e., functionally equivalent), even if structurally not equivalent to the disclosed structure performing the functions of the exemplary aspects, one or more embodiments, or the technology disclosed herein. Furthermore, while specific features of the disclosed technology may have been described above with respect to only one or more of the illustrated aspects or embodiments, such features may be combined with one or more other features of other embodiments, as may be necessary and advantageous for any given or particular application.

[0160] While the embodiments discussed herein relate to the systems and methods discussed above, these embodiments are intended to be exemplary and are not intended to limit the applicability of these embodiments to only those discussed herein. Although the invention has been described by way of examples, and although the embodiments have been described in considerable detail, the applicant does not intend to limit the scope of the appended claims or in any way to such details. Additional advantages and modifications will readily occur to those skilled in the art. Therefore, the invention, in its broader aspects, is not limited to the specific details shown and described, representative apparatuses and methods, and illustrative examples. Thus, deviations from these details may be made without departing from the spirit or scope of the applicant's overall inventive concept.

[0161] Example

[0162] The following is a non-exhaustive list of exemplary embodiments according to aspects of this disclosure.

[0163] 1. A radiographic imaging device, comprising:

[0164] Radiation sources used to emit radiation;

[0165] A radiation detector is configured to receive radiation from the radiation source and generate radiation data, wherein the radiation data includes a primary component and a scattered component.

[0166] The data processing system is configured as follows:

[0167] Receive the radiation data;

[0168] Based on the radiation data, a non-scattering corrected image is generated using a first data processing technique;

[0169] Estimate the scattering components of the radiation data;

[0170] Based on the scattering estimation, a scattering-only image is generated using a second data processing technique, wherein the second data processing technique differs from the first data processing technique; and

[0171] An image is generated based on the non-scattering corrected image and the scattering-only image.

[0172] 2. The imaging device according to Embodiment 1, wherein:

[0173] The radiation source includes a rotating X-ray source that emits a radiation beam;

[0174] The radiation detector includes an x-ray detector positioned to receive the radiation from the x-ray source; and

[0175] The device also includes:

[0176] A beamformer is configured to adjust the shape of the radiation beam emitted by the x-ray source such that a major section of the x-ray detector is directly exposed to the radiation beam, and at least one shadow section of the x-ray detector is blocked by the beamformer from being directly exposed to the radiation beam.

[0177] 3. The imaging apparatus according to any one of Embodiments 1 and 2, wherein the scattering component of the radiation data is estimated based on scattering data measured in the at least one shadow zone.

[0178] 4. The imaging apparatus according to any one of embodiments 1 to 3, wherein:

[0179] Generating the nonscattering corrected image includes reconstructing the radiation data; and

[0180] Generating the scatter-only image includes:

[0181] A scattering correction image is reconstructed based on the radiation data and the scattering estimation; and

[0182] Subtract the non-scattering corrected image from the scattering corrected image.

[0183] 5. The imaging apparatus according to any one of embodiments 1 to 3, wherein:

[0184] Generating the nonscattering corrected image includes reconstructing the radiation data, wherein the first data processing technique includes a kernel with a higher resolution than the second data processing technique; and

[0185] Generating the scatter-only image involves using a stripe artifact reduction algorithm to reconstruct the scattering estimate.

[0186] 6. The imaging apparatus according to any one of embodiments 1 to 3, wherein:

[0187] Generating the nonscattering corrected image includes reconstructing the radiative data using a first grid; and

[0188] Generating the scatter-only image includes:

[0189] The scattering-only image is reconstructed using a second grid with a reconstruction time that is faster than that of the first grid;

[0190] The reconstructed scatter-only image is resampled using the first grid.

[0191] The scattering-only image is processed based on a third data processing technique, wherein the third data processing technique is determined based on the non-scattering-corrected image.

[0192] 7. The imaging apparatus according to any one of embodiments 1-6, wherein the first data processing technique includes a high-resolution kernel and the second data processing technique includes a smoothing kernel.

[0193] 8. The imaging apparatus according to any one of embodiments 1-6, wherein the second data processing technique is determined based on the nonscattering corrected image.

[0194] 9. The imaging apparatus according to any one of embodiments 1-8 further includes processing the scattering-only image based on a third data processing technique, wherein the third data processing technique is determined based on the non-scattering corrected image.

[0195] 10. The imaging apparatus according to Embodiment 9, wherein the third data processing technique includes a Gaussian filter that uses voxel differences in the nonscattering corrected image to determine kernel weights for the third data processing technique.

[0196] 11. The imaging apparatus according to any one of embodiments 9 and 10, wherein processing the scatter-only image includes processing the scatter-only image using anisotropic differential filter parameters obtained from the non-scattering corrected image.

[0197] 12. The imaging apparatus according to any one of embodiments 1-11, wherein the first data processing technique is applied to the radiation data before the reconstruction of the non-scattering corrected image, and the second data processing technique is applied to the scattering estimation before the reconstruction of the scattering-only image.

[0198] 13. A method for generating a radiographic image, comprising:

[0199] Receive radiation data from a radiation imaging device, wherein the radiation data includes a principal component and a scattering component;

[0200] Based on the radiation data, a non-scattering correction dataset is generated using a first data processing technique;

[0201] Estimate the scattering components of the radiation data;

[0202] Based on the scattering estimation, a scattering-only dataset is generated using a second data processing technique, wherein the second data processing technique differs from the first data processing technique; and

[0203] Images are generated based on the non-scattering correction dataset and the scattering-only dataset.

[0204] 14. The method according to Example 13, wherein:

[0205] Generating the nonscattering correction dataset based on the radiation data and using the first data processing technique includes:

[0206] Generate the nonscattering correction line integral; and

[0207] The nonscattering correction line integral is reconstructed and processed using the first data processing technique to generate a nonscattering correction image;

[0208] Generating the scatter-only dataset based on the scattering estimation and using a second data processing technique includes:

[0209] Generate only the scattering line integral; and

[0210] The reconstructed scatter-only line integral is reconstructed and processed using the second data processing technique to generate a scatter-only image; and

[0211] Generating the image based on the non-scattering correction dataset and the scattering-only dataset includes adding the scattering-only image to the non-scattering correction image.

[0212] 15. The method according to any one of Embodiments 13 and 14, wherein the second data processing technique is based on the nonscattering corrected image.

[0213] 16. The method according to Example 13, wherein:

[0214] Generating the nonscattering correction dataset based on the radiation data and using the first data processing technique includes:

[0215] Generate the nonscattering correction line integral; and

[0216] The reconstructed nonscattering correction line integral is reconstructed and processed using the first data processing technique to generate a nonscattering correction image;

[0217] Generating the scatter-only dataset based on the scattering estimation and using the second data processing technique includes:

[0218] Generate the scattering correction line integral;

[0219] Reconstruct the integral of the scattering correction line;

[0220] Determine the difference between the reconstructed nonscattering correction line integral and the reconstructed scattering correction line integral; and

[0221] The difference is processed using the second data processing technique to generate a scatter-only image; and

[0222] Generating the image based on the non-scattering correction dataset and the scattering-only dataset includes adding the scattering-only image to the non-scattering correction image.

[0223] 17. The method according to Example 13, wherein:

[0224] Generating the nonscattering correction dataset based on the radiation data and using the first data processing technique includes:

[0225] Generate the nonscattering correction line integral; and

[0226] The nonscattering correction line integral is reconstructed and processed using the first data processing technique to generate a nonscattering correction image;

[0227] Generating the scatter-only dataset based on the scattering estimation and using the second data processing technique includes:

[0228] Generate only the scattering line integral; and

[0229] The second data processing technique is used to reconstruct and process the scatter-only line integral to generate a scatter-only image; and

[0230] Generating the image based on the non-scattering correction dataset and the scattering-only dataset includes adding the scattering-only image to the non-scattering correction image.

[0231] 18. The method according to Example 13, wherein:

[0232] Generating the nonscattering correction dataset based on the radiation data and using the first data processing technique includes:

[0233] Generate the nonscattering correction line integral; and

[0234] The nonscattering correction line integral is reconstructed and processed using the first data processing technique to generate a nonscattering correction image;

[0235] Generating the scatter-only dataset based on the scattering estimation and using the second data processing technique includes:

[0236] Generate the scattering correction line integral;

[0237] Reconstruct the scattering correction line integral to generate a scattering correction image;

[0238] Determine the difference between the reconstructed nonscattering correction line integral and the reconstructed scattering correction line integral; and

[0239] The difference is processed using the second data processing technique to generate a scatter-only image; and

[0240] Generating the image based on the non-scattering correction dataset and the scattering-only dataset includes adding the scattering-only image to the non-scattering correction image.

[0241] 19. The method according to Example 13, wherein:

[0242] Generating the nonscattering correction dataset based on the radiation data and using the first data processing technique includes:

[0243] Generate the nonscattering correction line integral; and

[0244] The non-scattering correction line integral is processed using the first data processing technique;

[0245] Generating the scatter-only dataset based on the scattering estimation and using the second data processing technique includes:

[0246] Generate only the scattering line integral; and

[0247] The scatter-only line integral is processed using the second data processing technique; and

[0248] Generating the image based on the non-scattering correction dataset and the scattering-only dataset includes:

[0249] The main data line integral is separated based on the difference between the processed non-scattering correction line integral and the processed scattering-only line integral;

[0250] The main data line integral is reconstructed to generate the image.

[0251] 20. A radiotherapy delivery device, comprising:

[0252] A rotatable gantry system, which is positioned at least partially around a patient support;

[0253] A first radiation source coupled to the rotatable gantry system, the first radiation source being configured as a therapeutic radiation source;

[0254] A second radiation source coupled to the rotatable gantry system, the second radiation source being configured as an imaging radiation source having an energy level lower than that of the therapeutic radiation source;

[0255] A radiation detector, coupled to the rotatable pedestal system and positioned to receive radiation from the second radiation source; and

[0256] The data processing system is configured as follows:

[0257] Receive radiation data, wherein the radiation data includes a primary component and a scattered component;

[0258] Based on the radiation data, a non-scattering correction dataset is generated using a first data processing technique;

[0259] Estimate the scattering components of the radiation data;

[0260] Based on the scattering estimation, a scattering-only dataset is generated using a second data processing technique, wherein the second data processing technique is different from the first data processing technique.

[0261] An image is generated based on the non-scattering correction dataset and the scattering-only dataset; and

[0262] During adaptive IGRT, a dose of therapeutic radiation is delivered to the patient via the first radiation source based on the image.

Claims

1. A radiographic imaging device, comprising: Radiation sources used to emit radiation; A radiation detector is configured to receive radiation from the radiation source and generate radiation data, wherein the radiation data includes a primary component and a scattered component. The data processing system is configured as follows: Receive the radiation data; Based on the radiation data, a non-scattering corrected image is generated using a first data processing technique; The scattering components are estimated based on the radiation data; Based on the estimated scattering components, a scattering-only image is generated using a second data processing technique, wherein the second data processing technique is different from the first data processing technique; The scattering-only image is processed using a noise-suppressing data processing technique that is stronger than the noise-suppressing data processing technique applied to the non-scattering corrected image; as well as A final scatter-corrected radiometric image is generated based on the non-scatter-corrected image and the scatter-only image.

2. The imaging apparatus according to claim 1, wherein: Generating the nonscattering corrected image includes reconstructing the radiation data; as well as Generating the scatter-only image includes: A scattering correction image is reconstructed based on the radiation data and the estimated scattering components; as well as Subtract the non-scattering corrected image from the scattering corrected image.

3. The imaging apparatus according to claim 1, wherein: Generating the nonscattering corrected image includes reconstructing the radiation data, wherein the first data processing technique includes a kernel with a higher resolution than the second data processing technique; as well as Generating the scatter-only image involves using a stripe artifact reduction algorithm to reconstruct the estimated scattering components.

4. The imaging apparatus according to claim 1, wherein: Generating the nonscattering corrected image includes reconstructing the radiation data using a first grid; as well as Generating the scatter-only image includes: The scattering-only image is reconstructed using a second grid with a reconstruction time that is faster than that of the first grid; The reconstructed scatter-only image is resampled using the first grid. The scattering-only image is processed based on a third data processing technique, wherein the third data processing technique is determined based on the non-scattering-corrected image.

5. The imaging apparatus of claim 1, wherein the first data processing technique includes a high-resolution kernel, and the second data processing technique includes a smoothing kernel; or The second data processing technique is determined based on the nonscattering corrected image.

6. The imaging apparatus of claim 1, further comprising processing the scattering-only image based on a third data processing technique, wherein the third data processing technique is determined based on the non-scattering corrected image.

7. The imaging apparatus of claim 6, wherein the third data processing technique comprises a Gaussian filter that uses voxel differences in the nonscattering corrected image to determine kernel weights for the third data processing technique.

8. The imaging apparatus of claim 6, wherein processing the scatter-only image includes processing the scatter-only image using anisotropic differential filter parameters obtained from the non-scattering corrected image.

9. The imaging apparatus according to any one of claims 1-8, wherein the first data processing technique is applied to the radiation data prior to the reconstruction of the non-scattering corrected image, and the second data processing technique is applied to the estimated scattering component prior to the reconstruction of the scattering-only image.

10. A method for generating a radiographic image, comprising: Receive radiation data from a radiation imaging device, wherein the radiation data includes a principal component and a scattered component; Based on the radiation data, a nonscattering correction line integral is generated using a first data processing technique, and the nonscattering correction line integral is reconstructed. The scattering components are estimated based on the radiation data; A scattering correction line integral is generated based on the radiation data and the estimated scattering components, and the scattering correction line integral is reconstructed. By subtracting the reconstructed non-scattering correction line integral from the reconstructed scattering correction line integral, a second data processing technique is used to generate a scattering-only line integral, wherein the second data processing technique is different from the first data processing technique. The scattering-only line integral is processed using a noise-suppressing data processing technique that is more powerful than the noise-suppressing data processing technique applied to the non-scattering correction line integral. The final scattering correction line integral is generated by adding the scattering-only line integral to the reconstructed non-scattering correction line integral and the scattering-only line integral. as well as The final CT image is reconstructed based on the integral of the final scattering correction line.

11. A radiotherapy delivery device, comprising: A rotatable gantry system, positioned at least partially around a patient support; A first radiation source coupled to the rotatable gantry system, the first radiation source being configured as a therapeutic radiation source; The radiographic imaging apparatus according to any one of claims 1-9, wherein the radiation source for emitting radiation is a second radiation source, the second radiation source being configured as an imaging radiation source having an energy level lower than that of the therapeutic radiation source, wherein the radiation detector is positioned to receive radiation from the second radiation source, and wherein the second radiation source and the radiation detector are coupled to the rotatable gantry system; and wherein, The data processing system is also configured to: During adaptive IGRT, a dose of therapeutic radiation is delivered to the patient via the first radiation source based on the image.

Images