X-ray computed tomography apparatus, method, and program

Abstract

To perform more accurate Automatic Exposure Control (AEC).SOLUTION: An X-ray computed tomography (CT) apparatus according to an embodiment includes a processing circuit. The processing circuit acquires projection data obtained from a first CT scan of an object and first CT image data reconstructed from the projection data. On the basis of a noise propagation model between the X-ray projection data and the CT image data, at least portions of the projection data, an imaging region-of-interest (ROI) of a second CT scan of the object, and a target image quality level of the imaging ROI are used as inputs to the noise propagation model to determine X-ray tube current modulation information for the second CT scan, and the second CT scan is performed on the basis of the X-ray tube current modulation information.SELECTED DRAWING: Figure 1A

Description

[Technical Field]

[0001] The embodiments disclosed in the present specification and drawings relate to an X-ray computed tomography apparatus, method, and program.

[0002] In particular, the present disclosure relates to a method, apparatus, and non-transitory computer-readable storage medium for fast patient-specific image region-of-interest noise control in AEC prediction. [Background technology]

[0003] In computed tomography (CT), a good way to reduce patient dose with desirable diagnostic image quality is by modulating the X-ray tube current, i.e., using the method of automatic exposure control (AEC). AEC has been widely used in routine clinical scanning for different protocols and anatomical structures of the body. However, the limited patient information available before a typical scan affects AEC prediction. Furthermore, patient variability, such as patient size, anatomical differences, and position during the scan, makes AEC prediction even more difficult. In addition, most image quality requirements are task-dependent and become more complex when using different protocols. Therefore, accurate AEC remains a challenging problem, especially for patient-specific scans.

[0004] Due to the limitations of pre-scanning, current AEC is usually based on two-dimensional radiographic images (typically one or two projection views). Therefore, it is difficult to obtain sufficient tomographic image information. In addition, simple model or look-up table methods using limited two-dimensional radiographic images cannot meet the requirements of task-dependent patient-specific AEC.

[0005] Most CT vendors offer AEC capabilities for clinical scans, but unfortunately, current methods make it difficult to perform AEC of patient-specific regions of interest due to limitations in simple models and pre-acquired patient information.

[0006] Recently, organ-based tube current modulation techniques have been developed to reduce radiation dose to sensitive organs. One technique, for example, utilizes machine learning techniques to provide coarse CT reconstruction, organ segmentation, and dose distribution estimation in real time. The technique uses this information to determine a tube current curve that minimizes patient risk indicators, such as effective dose, while maintaining consistent image quality.

[0007] The foregoing background discussion is intended to provide a general context for the present disclosure. The inventor's work, to the extent described in this background section, is not admitted expressly or impliedly as prior art to the present disclosure, as are aspects of this specification that are not admitted as prior art at the time of filing. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] U.S. Patent No. 9,706,972 [Patent Document 2] US Patent Application Publication No. 2019 / 0231296 [Patent Document 3] US Patent Application Publication No. 2016 / 0242712 [Patent Document 4] US Patent Application Publication No. 2015 / 0297165 [Patent Document 5] Patent No. 5670065 [Non-patent literature]

[0009] [Non-Patent Document 1] LIM et al., "Image quality and radiation reduction of 320-row area detector CT coronary angiography with optimal tube voltage selection and an automatic exposure control system: comparison with body mass index-adapted protocol", The International Journal of Cardiovascular Imaging 31,01 / 22 / 2015,12 Pages [Non-patent document 2] SOOKPENG et al., “Comparison of different phantom designs for CT scanner tube automatic current modulation system tests”, Journal of Radiological Protection, Volume 33, Number 4, 09 / 11 / 2013, Abstract Only, 5 Pages [Non-patent document 3] MUSSMANN et al., "Organ-based tube current modulation in chest CT. A comparison of three vendors",Radiography,Volume 27,Issue 1,February 2021,12 Pages [Non-patent document 4] KLEIN et al., “Patient-specific radiation risk-based tube current modulation for diagnostic CT”, Medical Physics, 04 / 14 / 2022, 28 Pages Summary of the Invention [Problem to be solved by the invention]

[0010] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to perform more accurate AEC. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be positioned as other problems. [Means for solving the problem]

[0011] An X-ray computed tomography (CT) apparatus according to an embodiment includes a processing circuit that acquires projection data obtained by a first CT scan of an object and first CT image data reconstructed from the projection data, determines X-ray tube current modulation information for the second CT scan based on a noise propagation model between the X-ray projection data and the CT image data using at least a portion of the projection data, a region-of-interest (ROI) imaging the object, and a target image quality level for the ROI as inputs to the noise propagation model, and performs the second CT scan based on the X-ray tube current modulation information. [Brief explanation of the drawings]

[0012] [Figure 1A] FIG. 1A is a flow diagram of an algorithm according to an exemplary embodiment described herein. [Figure 1B] FIG. 1B is a flow diagram of an algorithm according to an example embodiment described herein. [Figure 1C] FIG. 1C is a flow diagram of an algorithm according to an exemplary embodiment described herein. [Figure 1D] FIG. 1D is a flow diagram of an algorithm according to an example embodiment described herein. [Figure 2] FIG. 2 is a diagram illustrating an exemplary pre-scan according to an exemplary embodiment described herein. [Figure 3]FIG. 3 illustrates an example predicted AEC curve according to an example embodiment described herein. [Figure 4] FIG. 4 shows the processed pre-scan image. [Figure 5] FIG. 5 is a diagram illustrating an implementation of a radiography gantry included in a CT device or CT scanner, according to an exemplary embodiment of the present disclosure. [Figure 6] FIG. 6 is a block diagram of an apparatus used to implement a method according to an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0013] An aspect of the present disclosure is a method for performing X-ray computed tomography (CT) imaging, the method including: acquiring a first set of projection data obtained in a first CT scan of an object using a CT imaging device; acquiring first CT image data reconstructed from the acquired first set of projection data; determining X-tube current modulation information for the second CT scan of the object based on a noise propagation model between the X-ray projection data and the CT image data and using as input the acquired first set of projection data, information indicating an imaging region-of-interest (ROI) for the second CT scan, and a set target image quality level for the imaging ROI; and performing the second CT scan of the object based on the acquired X-tube current modulation information.

[0014] A further aspect of the present disclosure is a non-transitory computer-readable medium storing computer-executable instructions for causing a computer to perform a method for X-ray computed tomography (CT) imaging, the method including: acquiring image projection data from a pre-scan of an object using a CT device; performing analytical reconstruction of the acquired image projection data to obtain a reconstructed image; performing sparse sampling of image slices of the reconstructed image to generate a sampled image; setting a region of interest (ROI) within the sampled image; determining an automatic exposure control (AEC) curve based on the set target image quality, the set ROI, and the acquired image projection data using a noise propagation model; and performing a CT scan of the object based on the determined AEC curve.

[0015] A further aspect of the present disclosure is an X-ray imaging device that can include a processing circuit configured to acquire a first set of projection data obtained in a first CT scan of an object using a CT imaging device, acquire first CT image data reconstructed from the acquired first set of projection data, determine X-tube current modulation information for the second CT scan of the object based on a noise propagation model between the X-ray projection data and the CT image data and using as inputs the acquired first set of projection data, information indicating an imaging region of interest (ROI) for the second CT scan, and a set target image quality level for the imaging ROI, and perform the second CT scan of the object based on the acquired X-tube current modulation information.

[0016] The above summary of exemplary embodiments and the following detailed description thereof are merely exemplary aspects of the disclosed technology and are not intended to be limiting.

[0017] As used herein, the singular is defined as one or more. As used herein, the plural is defined as two or more. As used herein, the term "another" is defined as at least a second or more. As used herein, the terms "including" and / or "having" are defined as equivalent to "comprising, including, including" (i.e., open language). References throughout this specification to "one embodiment," "a particular embodiment," "embodiment," "implementation," "example," or similar terms mean that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment of the present disclosure. Thus, the appearances of such phrases in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments without limitation.

[0018] Computed tomography (CT) scanners can have automatic exposure control (AEC) systems that aim to maintain image quality for patients of various sizes while keeping dose as low as reasonably practicable. Such systems are also designed to maintain image quality as individual patient sizes and attenuation vary over the length of the system. Some systems are based on the desired noise level in the image.

[0019] The present disclosure describes methods for improving the speed of generating AEC curves or maintaining or improving the accuracy of generating AEC curves in main imaging scans (e.g., computed tomography (CT) scans, tomosynthesis scans, and VCT (X-ray volume CT) scans). The present disclosure also describes information processing devices including processing circuitry and / or computer instructions stored on a non-transitory computer-readable storage medium for implementing the above methods.

[0020] The disclosed method can perform task-dependent patient-specific AEC prediction using pre-scans, 3D patient information and noise propagation models, compared to simple model or look-up table methods using 2D radiographic images.

[0021] The disclosed method incorporates sparse sampling of image slices and analytical AEC calculation to achieve faster speeds compared to other patient-specific radiation risk-based methods. Sparse sampling of image slices uses only key slices, accelerating the speed at which the AEC curve is determined while still achieving high-quality image scans. The AEC prediction uses 3D patient information from a pre-scan, which simplifies the calculation of the AEC curve. Furthermore, the disclosed noise propagation model improves the accuracy of the determined AEC curve. The faster speed and improved accuracy of generating the AEC curve can simplify routine clinical scans and improve diagnostic efficiency. Embodiments include a fast, patient-specific, image-region-of-interest noise-controlled AEC prediction framework using a 3D pre-scan.

[0022] Hereinafter, an embodiment of an information processing method will be described in detail with reference to the accompanying drawings.

[0023] FIG. 1A is a flow diagram of a patient-specific AEC according to one embodiment of the present disclosure.

[0024] In step S102, a low-dose pre-scan is performed to acquire projection data. In one embodiment, the low-dose pre-scan may be approximately 50% or less of the normal scan dose. Figure 2 illustrates data acquired by an exemplary pre-scan. The pre-scan process is described in more detail below with reference to Figure 1B.

[0025] In step S104, a fast analytical reconstruction of the projection data is performed to obtain a reconstructed image containing anatomical information of the patient.

[0026] In step S106, the image slices of the reconstructed image are sparsely sampled, which reduces subsequent processing time while maintaining the accuracy of the AEC prediction. The process of sparse sampling is described in more detail below with reference to FIG. 1C.

[0027] In step S108, a region of interest (ROI) is automatically or manually selected or set using the reconstructed image.

[0028] In step S110, an AEC curve is determined based on a target image quality level, such as a noise standard deviation, the set ROI, and the projection data obtained in the pre-scan using the disclosed noise propagation model, as will be described in more detail below with reference to FIG. 3. FIG. 3 shows an exemplary determined AEC curve.

[0029] In step S112, a normal CT scan (not a pre-scan) is performed based on the determined AEC curve.

[0030] As shown in FIG. 1B, in step S122, the gain and electronic noise variations of the scanner system are measured or set.

[0031] In step S124, a prescan is performed to determine the prescan power I, as will be described in more detail below. 0,p and prescan average count λ p is obtained.

[0032] In one embodiment, the patient-specific image region-of-interest noise-controlled AEC prediction framework can include sparse sampling of image slices. Figure 1C is a flow diagram of the sparse sampling of step S106. In this process, slice-by-slice sampling may not be necessary; a subset of the major slices may be sufficient. While processing slice-by-slice provides accurate AEC curves for each slice, it is not efficient. For example, in the processed pre-scan image (see Figure 4) and other patient scans, the shape variation of the whole body, especially the abdominal and pelvic regions, is not large.

[0033] In S106, sparse sampling is applied to the 3D pre-scan image, which reduces processing time. Note that prior to the sparse sampling step, a pre-scan may be performed using a low-dose helical whole-body scan in step S102.

[0034] Specifically, in step S132 of Figure 1C, soft tissue and bone regions are segmented from the reconstructed CT volume. Image segmentation is performed to generate a segmented image B' seg is generated. Segmentation methods include, but are not limited to, HU-based thresholding, which operates to segment soft tissue regions based on a range of HU values. Such segmentation can also be used to segment bone regions. For example, at least three types of tissue can be identified: soft tissue, lung tissue, and bone. Alternatively, other types of tissue (e.g., fat) can be identified, as well as other organ types such as breast, eye, reproductive organs, kidney, heart, liver, pancreas, and stomach. Other segmentation methods may also be used.

[0035] In step S134, each of the two component (eg, soft tissue and bone) volumes may be downsampled along the z-axis.

[0036] In step S136, the downsampling factor may be adjusted according to the pitch of the reconstructed slice.

[0037] In step S138, vectors for slice-by-slice pixel summation are generated for both the soft tissue volume and the bone volume.

[0038] In step S140, the gradients of the vector representing the change in the soft tissue region and the vector representing the change in the bone region are calculated.

[0039] In step S142, sparse slice selection is performed based on the gradient of the vector.

[0040] In contrast, a typical reconstructed image has a narrow image width and contains abundant information for generating an AEC curve. To improve efficiency, an adaptive sparse sampling method for image slices was developed based on the gradient information of the soft tissue and bone in the image. Specifically, the reconstructed image is segmented into two-component images of soft tissue and bone using a segmentation method such as a simple HU thresholding method. For each segmented two-component (soft tissue and bone) volume, each image slice is summed to generate a vector along the z-axis, which partially reflects the attenuation change from slice to slice. Furthermore, the gradient of the (soft tissue and bone) vector is obtained. This is calculated for every N slices (N can be determined based on actual case studies). A gradient threshold is set to select suitable slices for further processing. All slices obtained by soft tissue and bone segmentation are combined to obtain the final slice list.

[0041] Returning to step S108, in clinical applications, physicians are typically interested in certain regions of a patient obtained by a CT scan. The region of interest can be a single organ, such as the liver, or a region including multiple organs, such as the abdominal region. The region of interest can be manually selected from a pre-scan image or obtained by automatic segmentation. That is, the region of interest can be set using the pre-scan image. The region of interest can also be set based on user input or an anatomical detection process.

[0042] 1D is a flow diagram of the AEC curve determination in step S110. As shown in FIG. 1D, in step S152, a noise standard deviation is calculated based on a region of interest (ROI). In step S154, the region of interest, noise standard deviation, scanner system gain, electronic noise variation, vector, reconstruction operator, pre-scan power, and pre-scan average count are obtained. In step S156, an AEC curve is generated based on a noise propagation model using the projection data and the information obtained in step S154.

[0043] In one embodiment, a patient-specific image region-of-interest noise-controlled AEC prediction framework includes an analytical reconstruction-based noise propagation model between projection data and an image region of interest for AEC calculation. Image quality of the region of interest, especially the signal-to-noise ratio, is important for improved diagnosis. The corresponding noise standard deviation can be used in the process of determining the AEC curve.

[0044] (Measurement relationship between pre-scan and normal scan) The general equation for the detector measurement (c) and the X-ray tube source is:

number

[0045] where I0 and S(e) are the power and spectrum of the X-ray tube source, and ∫ e S(e)=1;(μ flt,l flt ) and (μ pat ,l pat ) are the attenuation and the filter and patient path lengths, respectively. If the normal scan has the same scan geometry and protocol as the prescan, the average measured counts per pixel of the detector (λ n,(u,v) ) can be calculated using the following formula:

number

[0046] In the formula, (λ p,(u,v) ) is the prescan average count, and I 0,p is the pre-scan power (see step S124 in FIG. 1B).

[0047] (Noise model with analytical reconstruction) Provided is the analytical reconstruction operator that maps the projection data y to the image x, namely:

number

[0048] The covariance of x can be calculated using the following formula:

number

[0049] where B is a linear operator and B' is the transpose of B. Once a region of interest (Ω) is selected, the size of the region of interest |Ω| is determined and the noise standard deviation s Ω can be calculated by the following formula:

number

[0050] (Data noise model) Instead of generating projection images by prescanning, a noise model can be constructed using Poisson and Gaussian noise to calculate the noise propagation from projection data to the reconstructed image. Generally, when the count level is above a predetermined threshold, the noise in the measured X-ray photon counts can be modeled using Poisson and Gaussian distributions. In such a case, the variability of the measured counts (c) can be expressed by the following equation:

number

[0051] where a is the gain of the scanner system and σ e 2 represents the variability of electronic noise (see step S122 in FIG. 1B). If the count level is below a predetermined threshold, an alternative, more accurate model may be used. In one embodiment, the noise model may be switched based on the target count level.

[0052] As represented in the noise model, the photon noise and electronic noise follow Poisson and Gaussian distributions. The electronic noise maintains the same distribution because it is determined by the Data Acquisition System (DAS) regardless of the measured count level. The photon noise can be simulated using a Gaussian distribution when high photon counts are measured, which can further simplify the model. At low count levels, a Poisson distribution more accurately represents the photon noise. A switching method can be designed to switch the calculation method for the photon noise based on the measured count level.

[0053] CT reconstruction usually uses the logarithm of the measurement, i.e., y. In terms of the relationship between the pre-scan and normal scan counts, the variability of the normal scan data is given by:

number

[0054] Normal Scan Power I 0,n is an unknown variable. 0,n is used to generate the AEC curves and is a per-view value in the projection domain. Therefore, the noise standard deviation s Ω can be calculated by the following formula:

number

[0055] (current modulation) Var(y n ) and s Ω Combining the above equations for , defines the relationship between the image region of interest and the noise standard deviation of the X-ray tube power for a typical scan. Once the noise standard deviation is preset, the I 0,n is the only unknown variable.

[0056] Current modulation can be performed along the z-direction or the xyz-direction. The z-direction, or z-axis, is the direction of the bore of the system (shown in FIG. 5) "into the page" at a location corresponding to each of the pre-scan views. As used herein, the term "longitudinal" refers to the "z" direction into the bore in the direction of the axis of rotation RA shown in FIG. 5. For modulation in the z-direction only, I is considered constant for all projection views per image slice. 0,n is calculated.

[0057] In the case of modulation in the xyz directions, the actual X-ray tube current can only vary smoothly from view to view and can be represented by a known function (f(α)). f(α) can be preset as a sine or cosine wave, where α is the projection angle. Then, I 0,n is calculated to generate the AEC curve for a typical scan. A simple example is that the AEC curve for one image slice follows a sine wave, i.e., sin(α Io,vi ), where vi is the index of the projection view with the projection angle α. The new AEC curve is0,n,sin sin(α Io,vi ) Assuming that the total photons in an image slice do not change between z-modulation and xyz-modulation, the following equation must be satisfied:

number

[0058] In this equation, there is one unknown variable: I 0,n,sin Therefore, it is easy to find the solution. Note that this method is also general to more advanced functions of f(α).

[0059] Referring again to FIG. 1A, in S102, the pre-scan power I 0,p and the pre-scan average count λ is obtained. As mentioned above, the gain a of the scanner system and the electronic noise variation σ e 2 is a scanner parameter and is measured by the scanner system.

[0060] In S108, in this example, the region of interest (Ω), the noise standard deviation (s Ω ) is set.

[0061] In S110, Var(y n ) and s Ω The AEC curve is determined by solving the equation for i is the i-th unit vector and B is the analytical operator (both are known). The unknown variables are I 0,n are only determined to generate the AEC curve.

[0062] The patient-specific imaging protocol method according to one embodiment of the present disclosure, as described above, can be implemented to apply to data acquired by a CT device or scanner. FIG. 5 illustrates an implementation of a radiography gantry included in a CT device or scanner. As shown in FIG. 5, the radiography gantry 1150 is depicted from a side view and further includes an X-ray tube 1151, an annular frame 1152, and a multi-row or two-dimensional array X-ray detector 1153. The X-ray tube 1151 and the X-ray detector 1153 are mounted on the annular frame 1152 diametrically opposite to the object OBJ, and the annular frame 1152 is rotatably supported about a rotation axis RA. A rotation device 1157 rotates the annular frame 1152 as fast as 0.4 seconds per rotation, while the object OBJ is moved along the axis RA toward or away from the illustrated plane.

[0063] Hereinafter, an embodiment of an X-ray CT apparatus according to the present disclosure will be described with reference to the accompanying drawings. It should be noted that X-ray CT apparatuses include various types of apparatuses, such as a rotating / rotating type apparatus in which both the X-ray tube and the X-ray detector rotate around the object to be inspected, and a fixed / rotating type apparatus in which multiple detector elements are arranged in a circular or horizontal pattern and only the X-ray tube rotates around the object to be inspected. The present disclosure is applicable to either type. Here, the rotating / rotating type, which is currently mainstream, will be exemplified.

[0064] The multi-slice X-ray CT apparatus further includes a high-voltage generator 1159, which generates a tube voltage applied to the X-ray tube 1151 through a slip ring 1158 to cause the X-ray tube 1151 to generate X-rays. The X-rays are emitted toward an object OBJ, whose cross-sectional area is represented by a circle. For example, the X-ray tube 1151 has an X-ray energy such that the average X-ray energy during a first scan is smaller than the average X-ray energy during a second scan. Thus, two or more scans corresponding to different X-ray energies can be obtained. An X-ray detector 1153 is disposed on the opposite side of the object OBJ from the X-ray tube 1151 to detect the emitted X-rays that have propagated through the object OBJ. The X-ray detector 1153 may further include individual detector elements or detector units and may be a photon-counting detector. In a fourth-generation geometric system, the X-ray detector 1153 may be one of multiple detectors arranged around the object OBJ in a 360° configuration.

[0065] The CT apparatus further includes other devices for processing detection signals from the X-ray detector 1153. The data acquisition circuit or DAS 1154 converts the signal output from the X-ray detector 1153 for each channel into a voltage signal, amplifies the signal, and further converts the signal into a digital signal. The X-ray detector 1153 and the DAS 1154 are configured to process a predetermined total number of projections per rotation (TPPR).

[0066] The above-mentioned data is transmitted through a non-contact data transmitter 1155 to a pre-processing device 1156 housed in a console external to the radiography gantry 1150. The pre-processing device 1156 performs certain corrections, such as sensitivity correction, on the raw data. A memory 1162 stores the resulting data, also called projection data, just before reconstruction processing. The memory 1162, along with a reconstruction device 1164, an input device 1165, and a display 1166, are connected to a system controller 1160 through a data / control bus 1161. The system controller 1160 controls a current regulator 1163, which limits the current to a level sufficient to drive the CT system. In one embodiment, the system controller 1160 implements optimized scan acquisition parameters.

[0067] In any generation CT scanner system, the detector may rotate and / or be fixed relative to the patient. In one implementation, the CT system described above may be a combined third-generation and fourth-generation geometry system. In a third-generation system, the x-ray tube 1151 and x-ray detector 1153 are mounted diametrically on an annular frame 1152 and rotate around the object OBJ as the annular frame 1152 rotates about its rotation axis RA. In a fourth-generation geometry system, the detector is fixedly mounted around the patient and the x-ray tube rotates around the patient. In an alternative embodiment, the radiography gantry 1150 has multiple detectors arranged on an annular frame 1152, which is supported by a C-arm and a stand.

[0068] The memory 1162 can store measurements indicative of x-ray exposure at the x-ray detector 1153. Additionally, the memory 1162 can store specialized programs for performing CT image reconstruction, material decomposition, and PQR estimation methods, including those described herein.

[0069] The reconstruction device 1164 can perform the above-described methods described herein. The reconstruction device 1164 may implement reconstruction according to one or more optimized image reconstruction parameters. Furthermore, the reconstruction device 1164 can perform pre-reconstruction image processing, such as volume rendering and image subtraction, as needed.

[0070] Reconstruction pre-processing of the projection data performed by pre-processing device 1156 can include, for example, detector calibration, correcting for detector non-linearities, and polar effects.

[0071] Post-reconstruction processing performed by the reconstruction device 1164 may include filter generation and image smoothing, volume rendering, and image subtraction, as needed. The image reconstruction process may implement the optimal image reconstruction parameters derived above. The image reconstruction process may be performed using filtered backprojection, iterative image reconstruction, or stochastic image reconstruction.

[0072] The reconstruction device 1164 can use memory to store, for example, projection data, forward projection training data, training images, uncorrected images, calibration data and parameters, and computer programs. The reconstruction device 1164 can also include processing support for machine learning, which may include calculating a reference data set based on the acquired spatial distribution of soft tissue regions and generating a filter by performing all or part of a machine learning process using the projection data set as input data and the reference data set as training data. The application of machine learning, which may include the application of an artificial neural network, also enables the generation of one or more evaluation values ​​representative of image quality.

[0073] The reconfiguration device 1164 may be implemented individually by a single processor or in a network or cloud configuration of processors. The reconfiguration device 1164 may include a central processing unit (CPU) (processing circuitry) that may be implemented as discrete logic gates, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other complex programmable logic device (CPLD). FPGA or CPLD implementations may be coded in VHDL (Variable Hardware Description Language), Verilog, or any other hardware description language, and the code may be stored directly in electronic memory within the FPGA or CPLD or as a separate electronic memory. Furthermore, the memory 1162 may be a non-volatile memory such as a read-only memory (ROM), an EPROM, an EEPROM, or a flash memory. The memory 1162 can be a volatile memory such as static or dynamic random access memory (RAM), and a processor such as a microcontroller or microprocessor can be provided to manage the electronic memory and the interaction between the FPGA or CPLD and the memory. In one embodiment, the reconstruction device 1164 can include a CPU and a graphics processing unit (GPU) to process and generate the reconstructed image. The GPU can be a dedicated or integrated graphics card that shares resources with the CPU, or can be one of various artificial intelligence-focused types of GPUs, including NVIDIA Tesla and AMD FireStream.

[0074] Alternatively, the CPU of the reconstruction device 1164 may execute a computer program including a set of computer-readable instructions that perform the functions described herein, the program being stored in any of the non-transitory electronic memory and / or hard disk drives, CDs, DVDs, flash drives, or any other known storage media mentioned above. Furthermore, the computer-readable instructions may be provided as a utility application, a background daemon, or an operating system component, or a combination thereof, and executed in conjunction with a processor such as an Intel® XEON® processor or an AMD® OPTERON™ processor, and an operating system such as Microsoft® 10, UNIX®, SOLARIS®, LINUX®, Apple MAC-OS®, and other operating systems known to those skilled in the art. Furthermore, the CPU of the reconstruction device 1164 may be implemented as multiple processors working in parallel to execute instructions.

[0075] In one implementation, the reconstructed image can be displayed on a display 1166. The display 1166 can be an LCD display, a CRT display, a plasma display, an OLED, an LED, or any other display known in the art.

[0076] The memory 1162 may be a hard disk drive, a CD-ROM drive, a DVD drive, a FLASH drive, RAM, ROM, or any other electronic storage device known in the art.

[0077] FIG. 6 is a block diagram of a device used to implement a method according to an exemplary embodiment of the present disclosure. The device may be, for example, a workstation running an operating system such as the Ubuntu Linux OS, Windows, a version of Unix OS, or Mac OS. The device 600 may include one or more central processing units (CPUs) 650 having multiple cores. The device 600 may also include a graphics board 612 having multiple GPUs, each with GPU memory. The graphics board 612 may perform some of the mathematical operations of the disclosed machine learning methods. The device 600 includes a main memory 602, typically random access memory (RAM), that houses software executed by the CPU 650 and GPUs, and a non-volatile storage device 604 that stores data and software programs. Several interfaces may be provided for interacting with the device 600, including an I / O bus interface 610, input / peripheral devices 618 such as a keyboard, touchpad, mouse, etc., a display adapter 616, and one or more displays 608, as well as a network controller 606 that enables wired or wireless communication over a network 99. The interface, memory, and processor may communicate through a system bus 626. The device 600 includes a power supply 621, which may be a redundant power supply.

[0078] In some embodiments, device 600 may include a CPU and a graphics card by NVIDIA in which the GPU has multiple CUDA cores.

[0079] The terms "processor" or "processing circuit" used in the above description refer to circuits such as a CPU, a GPU, an ASIC, and a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)). If the processor is, for example, a CPU, the processor performs its functions by reading and executing a computer program stored in a storage circuit. On the other hand, if the processor is, for example, an ASIC, instead of storing a computer program in a storage circuit, the function is directly incorporated into the processor's circuit as a logic circuit. It should be noted that each processor in the embodiments is not limited to the case where each processor is configured as a single circuit, and one processor may be configured by combining multiple independent circuits to perform its function. Furthermore, multiple components in each figure may be incorporated into one processor to perform its function.

[0080] According to at least one of the embodiments described above, more accurate AEC can be performed.

[0081] Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as defined in the claims.

[0082] With respect to the above embodiment, the following supplementary notes are disclosed as one aspect and optional features of the invention. (Appendix 1) 1. A method of performing X-ray computed tomography (CT) imaging, comprising: acquiring a first set of projection data obtained by a first CT scan of the object with a CT imaging device, and acquiring first CT image data reconstructed from the acquired first set of projection data; determining x-ray tube current modulation information for the second CT scan of the object based on a noise propagation model between x-ray projection data and CT image data and using as input at least a portion of the acquired first set of projection data, information indicative of an imaging region-of-interest (ROI) for the second CT scan, and a target image quality level for the imaging ROI; performing the second CT scan of the object based on the acquired x-ray tube current modulation information; The method comprising: (Appendix 2) 2. The method of claim 1, further comprising using the first CT image data to set the imaging ROI for the second CT scan. (Appendix 3) 2. The method of claim 1, wherein the step of determining the X-ray tube current modulation information further includes inputting at least one parameter characteristic of data acquisition by the CT imaging device into the noise propagation model. (Appendix 4) The method of claim 1, further comprising: performing sparse sampling of the acquired first CT image data to generate second CT image data; and setting the imaging ROI in the second CT image data. (Appendix 5) 2. The method of claim 1, further comprising setting the imaging ROI based on an anatomical detection process. (Appendix 6) 2. The method of claim 1, wherein the determined tube current modulation information comprises a tube current as a function of projection angle. (Appendix 7) 2. The method of claim 1, further comprising setting the imaging ROI based on input from a user. (Appendix 8) The method of claim 1, further comprising determining a size of the imaging ROI, wherein the step of determining the X-ray tube current modulation information further comprises determining the X-ray tube current modulation information based on the determined size of the imaging ROI. (Appendix 9) 1. A non-transitory computer-readable medium storing computer-executable instructions for causing a computer to perform a method of X-ray Computed Tomography (CT) imaging, the method comprising: The method comprises: acquiring image projection data from a pre-scan of the object; performing an analytical reconstruction of the acquired image projection data to obtain a reconstructed image; performing sparse sampling of image slices of the reconstructed image to generate a sampled image; selecting a Region-Of-Interest (ROI) of the sampled image; determining an Automatic Exposure Control (AEC) curve based on a target image quality, the selected ROI, and the image projection data using a noise propagation model; performing a CT scan based on the determined AEC curve; The non-transitory computer-readable medium comprising: (Appendix 10) The step of performing sparse sampling comprises: segmenting soft tissue and bone regions within the reconstructed image; generating slice-by-slice pixel summation vectors for both the soft tissue and bone regions; Calculating the gradient of the generated vector; guiding slice selection based on the calculated gradient of the vector; and 10. The non-transitory computer-readable medium of claim 9, further comprising: (Appendix 11) The step of determining the AEC curve comprises: obtaining the CT system gain, noise variance, prescan power, and prescan count, which is a prescan scanner-based value; determining the AEC curve based on scanner-based values; 10. The non-transitory computer-readable medium of claim 9, further comprising: (Appendix 12) 10. The non-transitory computer-readable medium of claim 9, wherein determining the AEC curve further comprises determining the AEC curve based on a predetermined target noise standard deviation. (Appendix 13) 11. The non-transitory computer-readable medium of claim 10, wherein determining the AEC curve further comprises determining the AEC curve based on the generated vector. (Appendix 14) 10. The non-transitory computer-readable medium of claim 9, further comprising performing current modulation of the CT scan in an axial direction of a CT system. (Appendix 15) 10. The non-transitory computer-readable medium of claim 9, further comprising performing current modulation of the CT scan based on a projection angle. (Appendix 16) 1. An x-ray imaging device including a processing circuit, The processing circuitry acquiring a first set of projection data obtained by a first CT scan of the object using a computed tomography (CT) imaging device; and acquiring first CT image data reconstructed from the acquired first set of projection data; determining x-ray tube current modulation information for the second CT scan of the object based on a noise propagation model between x-ray projection data and CT image data and using as input at least a portion of the acquired first set of projection data, information indicative of an imaging region-of-interest (ROI) for the second CT scan, and a target image quality level for the imaging ROI; configured to perform the second CT scan of the object based on the acquired x-ray tube current modulation information. X-ray imaging equipment. (Appendix 17) 17. The X-ray imaging device of claim 16, wherein the processing circuitry is further configured to use the first CT image data to set the imaging ROI of the second CT scan. (Appendix 18) 17. The X-ray imaging device of claim 16, wherein in determining the X-ray tube current modulation information, the processing circuitry is further configured to input at least one parameter indicative of characteristics of data acquisition by a CT device into the noise propagation model. (Appendix 19) 17. The X-ray imaging device of claim 16, wherein the processing circuitry is configured to set the imaging ROI based on an anatomical detection process. (Appendix 20) 17. The X-ray imaging device of claim 16, wherein the processing circuitry is further configured to set the imaging ROI based on input from a user. [Explanation of symbols]

[0083] 1150 Radiography Gantry 1151 X-ray tube 1152 Annular Frame 1153 X-ray detector 1154 Data Acquisition Circuit (DAS) 1155 Contactless Data Transmitter 1156 Pretreatment Device 1157 Rotating Device 1158 slip ring 1159 High Voltage Generator 1160 System Controller 1161 Data / Control Bus 1162 memory 1163 Current Regulator 1164 Reconfiguration Device 1165 Input Devices 1166 Display

Claims

1. acquiring projection data obtained by a first CT scan of the object and first CT image data reconstructed from the projection data; determining x-ray tube current modulation information for the second CT scan based on a noise propagation model between projection data and CT image data using at least a portion of the projection data, a region-of-interest (ROI) imaging for a second CT scan of the object, and a target image quality level for the ROI imaging as inputs to the noise propagation model; performing the second CT scan based on the X-ray tube current modulation information; An X-ray computed tomography apparatus comprising a processing circuit.

2. The processing circuitry further uses the first CT image data to set an imaging ROI for the second CT scan.

2. The X-ray computed tomography apparatus according to claim 1.

3. the processing circuitry further uses at least one parameter characteristic of data acquisition by the X-ray computed tomography apparatus as an input to the noise propagation model to determine the X-ray tube current modulation information.

3. An X-ray computed tomography apparatus according to claim 1 or 2.

4. The processing circuitry further performs sparse sampling of the first CT image data to generate second CT image data, and sets the imaging ROI using the second CT image data.

3. An X-ray computed tomography apparatus according to claim 1 or 2.

5. The processing circuitry further sets the imaging ROI based on an anatomical detection process.

3. An X-ray computed tomography apparatus according to claim 1 or 2.

6. the x-ray tube current modulation information includes x-ray tube current as a function of projection angle; 3. An X-ray computed tomography apparatus according to claim 1 or 2.

7. The processing circuitry further sets the imaging ROI based on input from a user.

3. An X-ray computed tomography apparatus according to claim 1 or 2.

8. The processing circuitry further determines a size of the imaging ROI, and determines the X-ray tube current modulation information based on the size of the imaging ROI.

3. An X-ray computed tomography apparatus according to claim 1 or 2.

9. acquiring projection data obtained by a first CT scan of an object using an X-ray CT device and first CT image data reconstructed from the projection data; determining x-ray tube current modulation information for the second CT scan based on a noise propagation model between projection data and CT image data using at least a portion of the projection data, a region-of-interest (ROI) imaging of the second CT scan of the object, and a target image quality level for the ROI imaging as inputs to the noise propagation model; performing the second CT scan based on the x-ray tube current modulation information; An X-ray computed tomography method comprising:

10. acquiring projection data obtained by a first CT scan of an object using an X-ray CT device and first CT image data reconstructed from the projection data; determining x-ray tube current modulation information for the second CT scan based on a noise propagation model between projection data and CT image data using at least a portion of the projection data, a region-of-interest (ROI) imaging of the second CT scan of the object, and a target image quality level for the ROI imaging as inputs to the noise propagation model; performing the second CT scan based on the x-ray tube current modulation information; An X-ray computed tomography program that causes a computer to execute the above.

Images