Pixel determination method, X-ray CT apparatus, and program

Abstract

To determine an inferior pixel.SOLUTION: In a pixel determination method, a pixel with poor performance is determined from among pixels of a detector, based on a count image and a spectrum image generated based on multiple scans, and information on the determined pixel is stored.SELECTED DRAWING: Figure 11

Description

[Technical Field]

[0001] Embodiments disclosed herein and in the drawings relate to a pixel determination method, an X-ray CT apparatus, and a program. [Background technology]

[0002] In conventional X-ray CT scanners, energy-integrating detectors were used to measure CT projection data. Currently, photon-counting detectors (PCDs) are being effectively used as a replacement for energy-integrating detectors. In photon-counting CT systems, a semiconductor-based detector using direct conversion is used to decompose the energy of individual incident photons and generate multiple energy bin counts for each integration period.

[0003] In such sensor materials (e.g., CdTe / CZT / Si), photon energy is accumulated, i.e., a charge cloud is formed, which drifts toward the cathode under the applied electric field. The detector's energy response is largely degraded or distorted by charge sharing, K-escape, and scattering effects during the energy accumulation and subsequent charge induction processes, as well as by electronic noise from the associated front-end electronic circuitry. Due to the finite signal induction time, under high counting rate conditions, pulse pile-up also distorts the detector's energy response.

[0004] Due to the non-uniformity of the sensor material and the performance variability of the associated front-end electronic circuits, the actual detector responses of the integrated detector pixels differ slightly from one another. With proper calibration of the forward model, most pixels can generate measurements that yield good path length estimates by solving the inverse problem.

[0005] However, if the performance of a pixel far exceeds the nominal value, the measured value of that pixel may become unusable and needs to be excluded from subsequent processing. Different from the conventional scintillator-based Energy Integration Detector (EID), the following problems may occur in "bad" pixels in a photon counting detector. (1) Noisy counting background, (2) Abnormal counting performance, (3) Abnormal energy resolution, (4) Nonlinear energy response, (5) Incorrect energy threshold. As a result, the pixels may become uncalibratable and uncorrectable, and the screening criteria become more complex than those of the conventional EID system.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0007] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to determine defective pixels. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problems. The problems corresponding to the respective effects of each configuration shown in the embodiments described later can also be regarded as other problems.

Means for Solving the Problems

[0008] The pixel determination method according to the embodiment determines defective pixels among the pixels of the detector based on a counting image and a spectrum image generated based on a plurality of scans, and stores information on the determined pixels.

Brief Description of the Drawings

[0009] [Figure 1]Figure 1 shows an example of the response function Sb(E) of a PCD bin in a photon counting detector, with each curve representing an example of the function for each energy bin. [Figure 2] Figure 2 shows an example of an image affected by a faulty photon counting detector used during the image capture process. [Figure 3] Figure 3 is a flowchart showing the process for determining whether a counting-type photodetector is functioning correctly in terms of acquiring spectral and counting images. [Figure 4A] Figure 4A shows an example of a defective spectral pixel table before and after the defective pixel detection process shown in Figure 3. [Figure 4B] Figure 4B shows an example of a defective spectral pixel table before and after the defective pixel detection process shown in Figure 3. [Figure 5A] Figure 5A shows an example of a defective pixel count table before and after the defective pixel detection process shown in Figure 3. [Figure 5B] Figure 5B shows an example of a defective pixel count table before and after the defective pixel detection process shown in Figure 3. [Figure 6] Figure 6 is a schematic diagram of a calibration technique that performs a series of calibration scans, according to one embodiment of Figure 3. [Figure 7] Figure 7 shows the path lengths of the calibration slab at different detector pixels, which can be used as part of a series of calibration scans, according to one embodiment of Figure 3. [Figure 8] Figure 8 shows an example of a calibration slab configuration in which the calibration slab is aligned in the Z direction along at least some of the bed apparatus or other movable mechanisms in the detector field. [Figure 9A] Figure 9A shows an addition scheme that can be used for calibration and processing. [Figure 9B] Figure 9B shows an addition scheme that can be used for calibration and processing. [Figure 9C]Figure 9C shows an addition scheme that can be used for calibration and processing. [Figure 9D] Figure 9D shows an addition scheme that can be used for calibration and processing. [Figure 10] Figure 10 is a schematic diagram illustrating how different X-ray tube positions can result in different path lengths for slab scans during calibration. [Figure 11] Figure 11 shows a CT scanner system that can incorporate the technology of this disclosure. [Modes for carrying out the invention]

[0010] The following describes in detail the pixel determination method, the X-ray CT apparatus, and the program, with reference to the drawings.

[0011] Throughout this specification, the expression “a particular embodiment” or “one embodiment” means that any specific feature, structure, material, or property described in relation to that embodiment is included in at least one embodiment of this application, but not in all embodiments.

[0012] Therefore, phrases such as “in one embodiment” or “in one embodiment” appearing in various places throughout this specification do not necessarily refer to the same embodiment of this application. Furthermore, the particular features, structures, materials, or properties may be combined in any suitable manner in one or more embodiments.

[0013] This disclosure relates to a photon counting type CT scanner system for material discrimination, the CT scanner system comprising one or more X-ray tubes that emit X-ray radiation, and an array of detector pixels that receive X-ray radiation propagating through the field of view of the CT scanner system.

[0014] In transmission measurements using a photon counting energy-resolved detector (PCD), the forward model can be expressed by the following equation (1).

[0015]

Number

[0016] In the formula, S b (E) is the bin response function defined by the following formula (2).

[0017]

Number

[0018] R(E, E) is the detector response function, where E bL and E bH and are the low energy threshold and high energy threshold of each counting bin, respectively. Figure 1 shows an example of a model of a typical S b (E) function in the PCD. The long tail beyond the energy window is induced by charge sharing, K-escape, and scattering effects. The low energy tail is mostly due to the limit of the energy resolution caused by the associated electronic noise. N0 is the total X-ray flux from the air scan, μ m and l m and are the linear attenuation coefficient and path length of the m-th basic substance, respectively. w(E) is the normalized incident X-ray spectrum. In reality, neither w(E) nor S b (E) can be accurately known, but they can be combined as one term S wb (E) = w(E)S b (E) (hereinafter, this term is called the weighted bin response function). If S wb (E) can be calibrated using measured values, the problem of decomposition under low beam conditions can be solved.

[0019] Under high X-ray beam scan conditions (e.g., several percent pulse pile-up), further spectral distortion is introduced into the measured values by pulse pile-up. As one method of correcting the pile-up effect, a correction term may be added. The forward model is expressed as the following formula (3).

[0020]

number

[0021] In the formula, Pb is a parameterization based on the photon energy, the measured bin count, and the total count.

[0022] In counting image measurement, total counts are used instead of individual bin counts, and the forward model can be adjusted as shown in equation (4).

[0023]

number

[0024] In the equation, μ(Er) is the local attenuation coefficient, and the projection integral p = ∫μ(E,r)dl, S w (E) is the total count weighted bin response function here, and P(E,N tot ) are E, N tot This is a total count pile-up correction term that is a function of p.

[0025] Due to the manufacturing process of photon counting detector sensors, the performance of the front-end electronic circuitry, and the complexity of the implementation process, substandard pixels (also called "defective pixels") may exhibit problems such as high background counting, non-uniformity of counting, abnormal energy response, and instability. As a result, the pixel may either fail to provide valid measurements with a stable forward model, or it may have a forward model that is unresponsive to path length information. In either case, the pixel is excluded from scanning measurements or subsequent processing. As will be discussed in detail below, substandard pixels are not necessarily so inadequate as to be considered "defective." In fact, a substandard detector pixel may be "defective" for the first imaging protocol, but not for the next.

[0026] In counting mode, image quality is highly sensitive to the total count in each projection, and the energy information of the recorded photons is not required. In spectral mode, not only is the recorded count accurate, but the energy bin of each recorded photon is also important. Therefore, the criteria for defective pixels differ in different imaging modes. If defective pixels are not properly identified and corrected during data processing, the resulting image may show ring / band artifacts that negatively affect diagnosis, as shown in Figure 2 (where ring artifacts 200A and 200B are visible).

[0027] Furthermore, detectors generally degrade with cumulative use, and new defective pixels may occasionally appear. These new defective pixels need to be captured in order to maintain consistent image quality throughout the system's lifespan.

[0028] Figure 3 is a flowchart of the in-situ processing 300, which determines whether the counting type photodetector is operating correctly for acquiring spectral and counting images. After a series of faulty pixel calibration scans are performed in step 310, processing is performed on those scans to evaluate the usability of the detector for spectral and counting images. That is, based on the counting and spectral images generated from multiple scans, pixels with poor performance among the detector pixels are determined. In other words, in the X-ray CT apparatus according to the embodiment, the system controller 1160, which acts as a control unit as shown in Figure 11, performs multiple scans, and the reconstruction device 1164, which has the function of a determination unit, determines pixels with poor performance among the detector pixels based on the counting and spectral images generated from the multiple scans. For example, faulty pixels can be identified using a low-noise image obtained from a uniform calibration phantom.

[0029] As shown in Figure 3, the processing of spectral images and counting images is fundamentally independent and can be performed either serially or in parallel. In the case of parallel processing, each process can be performed on separate computer systems. As an example, pixels that are inferior to the counting image according to a first criterion are identified, and pixels that are inferior to the spectral image according to a second criterion different from the first criterion are identified.

[0030] For processing the availability of the detector for spectral imaging, after spectrally oriented data is acquired by calibration scan 310, control of the process moves to step 320S, where the spectrally oriented data is processed based on multiple tests. Generally, each time the test system performs a test in step 320S, control moves to step 330S to determine whether the detector is available under the conditions of that test. If the detector is available, control returns to step 320S if there are any remaining spectral tests. If there are no remaining spectral tests, control terminates (or controls are moved to counting tests).

[0031] If the spectral test fails in step 330S, control moves to step 340S, where the system isolates the defective pixels in the defect detector and corrects them. Furthermore, in step 340S, the system generates another calibration phantom spectral image with the defective pixels isolated and tests the image quality (IQ) obtained after the correction. If the image quality is sufficiently good, in step 360S, the system can update the table of defective spectral pixels (for example, the table shown in Figures 4A and 4B) that the system can use for subsequent spectral image processing. Then, control can return to step 320S in preparation for testing the next spectral processing.

[0032] If the image quality is deemed insufficient in step 350S, the processing criteria may need adjustment (details below), and control moves to step 370S. After that, it returns to step 320S in preparation for the remaining spectral tests.

[0033] After or concurrently with processing the availability of the detector for spectral imaging, the system processes the availability of the detector for counting imaging. This process begins in step 320C, where counting-oriented data is processed based on several tests. Generally, each time the test system performs a test in step 320C, control moves to step 330C to determine whether the detector is available under the conditions of that test. If the detector is available, control returns to step 320C if there are any remaining counting tests. If there are no remaining counting tests, control terminates.

[0034] If the counting test fails in step 330C, control moves to step 340C, where the system isolates the defective pixels in the defect detector and corrects them. Furthermore, in step 340C, the system generates another calibration phantom counting image with the defective pixels isolated and tests the image quality (IQ) obtained through the correction. If the image quality is sufficiently good, in step 360S, the system updates a table of defective counting pixels (for example, the table shown in Figures 5A and 5B) that the system can use for processing subsequent counting images. Control then returns to step 320C in preparation for testing the next counting process.

[0035] If the image quality is deemed insufficient in step 350C, the processing criteria may need adjustment, and control moves to step 370C. Then, it returns to step 320C in preparation for the remaining counting tests.

[0036] When performing a calibration scan for defective pixels in step 310, the system preferably performs a series of calibration scans, which include the operational scan settings necessary to identify abnormal detector responses for each pixel. These can be multiple scan sets in different protocols, including configurations such as kVp, mA, bowtie shape, collimation, rotation speed, and integration time. These scans may include any one or more combinations of (1) to (4) below.

[0037] (1) A series of air scans at different intensities (e.g., from low mA to high mA (Figure 6) or from high mA to low mA) to test the detector's counting characteristics. The bowtie filter can be removed in these scans to ensure more uniform X-ray beam exposure in the detection field.

[0038] (2) A series of scans using slabs of different materials and thicknesses (as shown in Figure 8) at different intensities (e.g., from low mA to high mA, or vice versa) to test the spectral response characteristics of the detector. That is, multiple scans at different attenuation path lengths. Normalized bin counts are used in this analysis to remove pixels showing abnormal spectral responses.

[0039] (3) Dark scan (without X-rays) to test the background of the detector count for noise evaluation. That is, a dark scan that does not use X-rays to measure the count background.

[0040] (4) A series of air scans over several hours or days, with multiple intensity settings ranging from seconds to minutes.

[0041] In addition to testing for defective pixels based on the criteria of steps 330C and 330S, it is also possible to test the criteria for the entire system. Several examples of system-wide criteria (counting or spectral) are described below. However, the test methods described herein are not limited to the criteria described herein.

[0042] (Criteria 1) For example, according to Criterion Example 1, the background count rate is too high (excessive) in all bin counts, i.e., N bi >N offset Pixels that exhibit this behavior can be considered defective.

[0043] (Criterion 2) In the example reference 2 used in step 330C, the system can evaluate its counting performance based on one or more arbitrary combinations of (a) and (b) below.

[0044] (a) The counting rate in a given X-ray beam is too low or too high (N i >N high (mA) or N i <N low (mA), (b)N i (mA) / N i (mA0)>r high (mA) or N i (mA) / N i (mA0) <r low (mA). In the formula, N i (mA) is the measurement count at a specific mA of X-ray flux, N i (mA0) is the measurement count at the lowest mA X-ray beam for the same pixel when the pulse pile-up effect is minimized.

[0045] (Criterion 3) In the exemplary criterion 3 used in step 330S, the system is N bi / N tot >T high (mA,μ j ,L j ) or N bi / N tot >T low (mA,μ j ,L j By determining whether the spectral response is abnormal, the spectral performance can be evaluated. In the above formula, N bi This refers to the individual bin count in slab or filter or mA measurement, N tot This is the total count, T high and T low This refers to the attenuation μ j, L jThese are two bin response thresholds for normalized bin counts in slab or filter or mA measurements. These bin response thresholds can be determined to be 5-10% lower or higher than the average value of all pixels that passed the system-wide criterion and the count criterion. That is, if the ratio of count variation to the average count is greater than the first threshold, the pixel is stored in the defective pixel table.

[0046] (Criterion 4) As further testing in steps 330C and 330S, measurements in counting mode and spectral mode may be performed after performing residual least squares error to select anomalous pixels after calibrating the forward model. A typical criterion is the calibrated forward model S = Σ(N bi j -f bi j (μ j L j )) 2This can be derived using the following formula: j is the slab path length index, i is the pixel index, and b is the index of the individual energy bin. The mean and standard deviation of S can be calculated for adjacent pixel groups, and any pixel with a standard deviation of S greater than x can be marked as a defective pixel (for example, x is between 3 and 5). That is, if the deviation from the count mean is greater than the second threshold, the pixel is saved in the defective count pixel table. The scan used to calibrate the forward model can be several air scans combined with multiple slab scans using known materials and thicknesses at different mA settings (this will be discussed in detail later in relation to Figure 6). This scan tests whether the forward model can be accurately calibrated using the selected forward model. This scan can be used alone or in combination with any of the defective pixel screening criteria 1-3 described above. For example, if a pixel fails the least squares error test, it may be classified as a defective pixel even if it meets criteria 1-3. If a pixel passes the least squares error test but clearly does not meet criteria 1-3, it may still be considered a defective pixel. If a pixel passes the least squares error test and only slightly fails to meet criteria 1-3, that pixel may still be considered a good pixel prior to the initial IQ test using a uniform phantom (Step 4). In other words, in the pixel determination method according to the embodiment, inferior pixels may be determined by determining whether multiple scans match a calibrated forward model.

[0047] (Criterion 5) Under low-flux conditions where pile-up effects are not significant (part of the noise test), the counting performance of the detector can be approximated by a Poisson distribution. In the case of an ideal Poisson distribution, the variance of the counts is equal to the mean of the counts. Pixels with significant noise can be identified using the ratio of the variance to the mean of the counts. An example of a criterion can be defined as shown in equation (5) below.

[0048]

number

[0049] In the formula, N i is the total count of pixels in the field of view i, which has a typical integration time of the main operation scan and a sample size large enough for such an evaluation. The test protocol needs to satisfy the low X-ray flux condition nπ < y. n is the incident count rate per pixel, τ is the effective count insensitive time, and y can be selected between 0.01 and 0.1, indicating that the pile-up event is less than 1% or 10% of the total incident count. R can be selected between 1.5 and 2 according to the value of y.

[0050] (Criterion 6) In all detection X-ray flux related criteria, when the size of the pixels at the edge (corner) is different from the size of the pixels at the center, the threshold can be defined separately for the edge (corner) and the center. In one example method, the value N offset , N high , N low obtained by normalizing the pixel area is used.

[0051] (Criterion 7) Depending on the various differences in the calibration scans used for the determination of defective pixels, the criteria for different scan protocols may also be different. One example is for the high X-ray flux protocol. If a pixel fails the initial counting performance test but passes the low X-ray flux counting performance test, it may be treated as a good pixel in the operational scan using low X-ray flux.

[0052] (Criterion 8) After correcting for temporal variations in the X-ray beam (using readings from a valid reference detector), the detector pixels should exhibit stable count outputs, including variations caused by statistical noise during scanning. The count values ​​detected by the detector drift over time from this constant value, and pixels with temporal stability problems can be identified by using any measurable metric. For example, evaluation of the maximum count deviation from the mean or median, and / or a white noise test, can be performed as part of a reproducibility or reliability test. That is, the multiple scans performed in the embodiment may be multiple reproducibility scans and / or reliability scans. If the metric exceeds the preset range, the pixel is identified as a defective pixel. Furthermore, the detected count at one acquisition time and the detected count at another acquisition time separated by a date and time must match within the statistical error. If the detected count at any acquisition time is significantly outside the preset range, the pixel is identified as a defective pixel.

[0053] In steps 340C and 340S, after defective pixels are identified based on predetermined criteria, the readings of those defective pixels are interpolated. The interpolation can be designed to be performed in a raw count measurement domain where each bin count is interpolated from multiple neighboring pixels (step 340C). The interpolation can also be designed to be performed in a raw sinogram domain where the estimated path length of these pixels is interpolated from neighboring pixels (step 340S).

[0054] In steps 340C and 370C, and steps 340S and 370S, the criteria for defective pixels can be tested and adjusted using a uniform phantom made of water or other tissue-simulating material with a common influence. That is, the multiple scans used in the pixel determination method according to the embodiment may be scans using a calibration phantom. After the initial criteria are set and the defective pixels are interpolated, a phantom image can be created based on the calibrated forward model. If a ring (as shown in Figure 2) appears in the phantom image due to the initial criteria, the initial criteria are tightened (for example, using a smaller x value in criterion 4) and the next iteration is performed. That is, in the pixel determination method according to the embodiment, the image is reconstructed based on the first determination criterion, and the defective pixels are determined based on whether or not the reconstructed image contains a ring formed by the defective pixels. This is repeated until the ring artifact is minimized. The criterion value is fixed to the value in the final iteration.

[0055] Although not illustrated, it is also possible to relax the criteria if no ring appears in the image. When the current criteria meet the image quality requirements, there may be room to relax one or more criterion parameters (for example, by using a larger x value in criterion 4), and the iteration of screening, interpolation, and phantom image processing can be repeated until a ring appears, after which the criterion parameters are fixed to the values ​​from the iteration immediately preceding the ring's appearance.

[0056] In steps 360S and 360C, the final defective pixel criteria are saved as at least one software table and used for scanning or processing the subject. That is, in the pixel determination method according to the embodiment, information of pixels determined to be inferior to the defective spectral pixel table or defective count pixel table is stored. In other words, the storage device 1162, acting as a storage unit, stores information of the determined pixels. Depending on the variation in the analysis results of the calibration scan, there may be multiple maps or tables of defective pixels for different scan protocols.

[0057] Figures 4A and 4B show the result of updating an existing spectral defective pixel table with additional defective pixels for protocol S2, when defective pixels already existed that were stored for protocol S1. Similarly, Figures 5A and 5B show the result of updating an existing counting defective pixel table with additional defective pixels for protocol C1, when two defective pixels already existed that were stored for protocols C1 and C2. Defective pixel calibration and generation can be performed periodically to ensure the acquisition of degraded pixels. That is, in the pixel determination method according to the embodiment, multiple scans may be performed periodically to determine the discrepancy in detector performance in the time domain. The map or table of defective pixels is updated when a new defective pixel is identified. In one embodiment, the defective pixel table is stored only in a system that performs a process to disable defective pixels (e.g., interpolation). Alternatively, to increase parallel processing within the system, a processor can be attached to the detector substrate and / or module to store a substrate-specific or module-specific defective pixel table. In this configuration, the detector substrate and / or module are programmed at the start of the scan to determine which protocol is being used, and to locally perform data correction (e.g., interpolation) based on the protocol. That is, in the pixel determination method according to this embodiment, a protocol for determining the performance of inferior pixels is stored. Furthermore, in this configuration, arbitrary additions (e.g., additions related to Figures 9B-9D) are performed on the detector substrate and / or module to reduce the amount of data that needs to be sent from the detector substrate and / or module to the system. Similarly, pixel-specific, module-specific, and detector substrate-specific attenuation corrections can also be performed using similar tables stored in the corresponding module or detector substrate.

[0058] Detector arrays are typically composed of small units using modular or tiled designs. By applying all screening criteria, the total number of defective pixels in a detector module can be obtained. If the number of defective pixels exceeds a predetermined value, or if the defective pixels form clusters that still affect image quality even after correction, the detector module is considered defective and requires replacement. For example, as shown in Figure 5B, the counted defective pixel table indicates the beginning of clusters that may require replacement of module 16 in the detector substrate 3.

[0059] In the case of a finely divided pixel PCD configuration, it is possible to generate standard-resolution and high-resolution images, as shown in Figures 9A to 9D, using different addition methods. While not limited to these, in the case of an N×N subpixel configuration, the basic pixel units for processing include 1×1, 1×N, N×1, and N×N (shown in Figures 9A to 9D, respectively). Next, a defective pixel analysis corresponding to these addition methods is performed, and a defective pixel map is generated for each method. A specific defective pixel map is used in the scanning process of the subject to determine which pixels to exclude from the scanning process.

[0060] Furthermore, each defective micropixel can be identified in relation to the macropixel to which it belongs when used in an addition technique, and defective micropixels can be excluded during the addition process in each combined macropixel. In this configuration, the system can also maintain a map of defective macropixels based on the number of defective micropixels within each macropixel and exclude corresponding macropixels from subsequent scans. For example, if there are five or more defects among the nine micropixels in the same 3x3 macropixel, that macropixel is classified as a defective macropixel (either in a specific type of scan or regardless of the test results of other criteria at the macropixel level, depending on how the system is configured).

[0061] In typical fan-beam coverage with third-generation CT, a flat slab can be used as part of the calibration scan in step 310, with slightly different actual path lengths intersecting the detector array, as shown in Figure 7. The actual path length L per detector pixel in these calibration scans... i This can be calculated using the following formula (6).

[0062]

number

[0063] In the formula, T is the thickness of the calibration slab, and θ is the thickness of the calibration slab. i This is the projection fan angle of the detector pixel i on a detector module blade (DMB) consisting of multiple rows and channels.

[0064] Furthermore, calibration using slabs of different thicknesses can be performed using a static scan setting without rotation. These slabs are large enough to cover the entire detector array and remain well horizontal while acquiring all data. In the case of thick slabs, the aperture of the CT gantry cannot cover the entire detector surface at the position of a single slab, but the slab position is adjustable, and multiple scans can be used to cover the entire detector surface. In another embodiment, calibration using slabs of different thicknesses may be performed in a scan configuration with rotation.

[0065] Further system variations associated with different rotational speeds (e.g., tube bundle, shadow of the anti-scatter grid (ASG), etc.) are captured by the air scan and reference detector and can be corrected accordingly in the airflow term N0 of the forward model. For example, air scans at each rotational speed may be performed before scanning the patient (subject) to calibrate ASG deflection and other beam path variations during rotation that induce variations in the incident X-ray beam intersecting the detector in various fields of view.

[0066] Refer to Figure 8. Various calibration slabs can be combined with each other in a direction along at least a portion of the length of the bed, forming an elongated "wedge-shaped" phantom. This allows each calibration path length to be detected without readjusting the phantom by moving the bed's position (regardless of the transport mechanism that carries the slabs), thus facilitating the calibration process.

[0067] To increase the combination of calibration path lengths for each slab configuration, the X-ray tube can be positioned at various locations while the slab is fixed horizontally in the X-Y plane, as shown in Figure 10. Figure 10 is a schematic diagram illustrating how various path lengths can be generated in the slab scan during this calibration depending on the position of multiple tubes. As an example, for a given slab thickness T, the fan angle φ i In a detector pixel i positioned with , if the X-ray tube is placed at different positions (-θ, 0, θ), the measurement path length can be determined by the following equations (7)-(9).

[0068]

number

[0069]

number

[0070]

number

[0071] In the equations, x1 = θ - φ and x2 = θ + φ. In one embodiment, a typical range of φ is 0 to 25 degrees, and θ can be selected between 20 and 60 degrees depending on the slab thickness spacing. By using this park-and-shoot method, the number of path length samples in most detector channels can be tripled, thereby significantly reducing the number of calibration slabs required to cover the same or wider path length range. The number of calibration samples can also be increased by placing the X-ray tubes in four or more positions, according to the same calculation method as above. In the case of a wide cone coverage system, the calibration path length must be calculated based on the projection angle in both the channel direction and the row direction.

[0072] In one embodiment, the slab is flat and kept horizontal during calibration. This is because it reduces (suppresses) uncertainty in the road length. In another embodiment, the slab does not necessarily need to be flat or horizontal, as long as the road length is known and adjusted. Also in one embodiment, each slab is made of one material. In yet another embodiment, it does not necessarily have to be made of one material. For example, the slab may be made of multiple materials. Examples of slab materials may include polypropylene, water, aluminum, titanium or copper, imitation microstructures, other polymers, stainless steel or other metals, k-edge materials, and various microstructure imitation materials.

[0073] Multiple path lengths can also be obtained from one or more materials by laying at least one slab flat in various ways. For example, multiple slabs can be lined up in close proximity along the patient bed and scanned as the patient bed moves within the radiography gantry. The multiple slabs may be of the same height and made of multiple materials, of the same height and made of the same material, or of multiple heights and made of multiple materials. In another embodiment, the slabs can be held and suspended within the field of view of the CT scanner (e.g., using a robotic arm). In yet another embodiment, the multiple slabs can be lined up offset in the Z direction between two adjacent slabs. As the slabs move in and out of the scanning field of view, multi-stage path length data can be obtained sequentially.

[0074] Needless to say, in one embodiment, the above-described technology can be applied to a CT device or CT scanner. Figure 11 shows an embodiment of a horizontal radiography gantry provided on a CT device or CT scanner. As shown in Figure 11, the radiography gantry 1150 (shown in a side view) comprises 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, rotatably supported around a rotation axis RA, in opposite positions on either side of the subject OBJ (e.g., patient). While the subject OBJ (e.g., patient) is moved along axis RA perpendicular to the illustrated page, a rotation unit 1157 rotates the annular frame 1152 at a high speed of 0.4 seconds per revolution.

[0075] An embodiment of the X-ray CT apparatus according to the present invention is described below with reference to the attached drawings. Note that there are various types of X-ray CT apparatuses, for example, rotary / rotating type apparatuses in which the X-ray tube and X-ray detector rotate simultaneously around the subject being examined, and fixed / rotating type apparatuses in which a number of detection elements are arranged in a ring or planar shape, and only the X-ray tube rotates around the subject being examined. The present invention is applicable to any of these types. Here, we will explain using the currently mainstream rotary / rotating type apparatus as an example.

[0076] This multislice X-ray CT scanner further includes a high-voltage generator 1159 that generates a tube voltage applied to the X-ray tube 1151 via a slip ring 1158 so that the X-ray tube 1151 generates X-rays. The X-ray detector 1153 is located on the opposite side of the X-ray tube 1151 across the subject OBJ and detects emitted X-rays that have passed through the subject OBJ (e.g., the patient). The X-ray detector 1153 further comprises individual detector elements or units and may be a photon counting type detector. In the fourth-generation shape system, the X-ray detector 1153 may be one of a plurality of detectors arranged in a 360-degree configuration around the subject OBJ (e.g., the patient).

[0077] Furthermore, the CT scanner is equipped with a device that processes the detection signals from the X-ray detector 1153. The data acquisition circuit, or data acquisition system (DAS) 1154, converts the signals output from the X-ray detector 1153 for each channel into voltage signals, amplifies those voltage signals, and then converts the amplified signals into digital signals. The X-ray detector 1153 and the DAS 1154 are configured to process a predetermined total number of projections per rotation (TPPR).

[0078] The aforementioned data is sent via a non-contact data transmitter 1155 to a preprocessor 1156 housed in an external console of the radiographigant 1150. The preprocessor 1156 applies corrections, such as sensitivity correction, to the raw data. A memory 1162 stores the resulting data (also called projection data immediately before the reconstruction process). The memory 1162, along with the reconstruction device 1164, input device 1165, and display 1166, is connected to a system controller 1160 via a data and / or control bus 1161. The system controller 1160 controls a current regulator 1163 to limit the current to a level sufficient to drive the CT system. In one embodiment, the system controller 1160 implements the aforementioned optimal scan acquisition parameters. The reconstruction device 1164 may include circuitry configured to perform the aforementioned methods, such as method 600.

[0079] The methods and systems described herein can be implemented using multiple techniques, but generally relate to imaging devices and / or processing circuits that perform the methods described herein. In one embodiment, the processing circuit is implemented as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a generic logic array (GAL), a programmable logic array (PAL), a circuit that can be a one-time programmable logic gate (e.g., a circuit using a fuse), or a circuit that can be a reprogrammable logic gate, or a combination thereof. Furthermore, the processing circuit may include embedded and / or external non-volatile computer-readable memory (e.g., RAM, SRAM, FRAM®, PROM, EPROM, and / or EEPROM) that stores computer instructions (binary executable instructions and / or interpreted computer instructions) that control the computer's processor to perform the processing described herein. The computer processing circuit may implement one or more processors, each supporting one or more threads and each having one or more cores.

[0080] Furthermore, embodiments of this disclosure are shown in the following insert.

[0081] (1) A method for determining a substandard pixel in a computed tomography (CT) system using a finely pixelated photon counting detector, the method comprising, but not limited to, (a) performing a plurality of in-situ calibration scans using the CT system; (b) determining that at least one pixel of at least one detector of the CT system is substandard based on a first reference set based on counting images and spectral images created by the plurality of calibration scans; and (c) storing the indices of at least one pixel of at least one detector of the CT system that has been determined to be substandard in a map of substandard pixels.

[0082] (2) The method of (1) wherein the CT system is used to perform multiple in-situ calibration scans, the method comprising, but not limited to, performing (a) multiple air scans at different intensities, (b) multiple scans at different attenuation paths, (c) X-ray-free dark scans for measuring the counting background, (d) noise scans, and (e) multiple repeatability and / or reliability scans.

[0083] (3) A method of either (1) or (2) that includes determining whether the plurality of scans match a calibrated forward model, but is not limited to determining whether the performance of at least one pixel of at least one detector of the CT system is inferior.

[0084] (4) A method of any one of (1) to (3) that determines that the performance of at least one pixel of at least one detector of the CT system is inferior based on the first reference set, but includes (b1) determining that the plurality of calibration scans indicate that the performance of at least one pixel of at least one detector of the CT system is inferior based on the first reference set, (b2) creating a second reference set different from the first reference set, and (b3) determining that the plurality of calibration scans indicate that the performance of at least one pixel of at least one detector of the CT system is inferior based on the second reference set.

[0085] (5) A method which is any one of (1) to (4), and which includes performing scans using a calibration phantom, but is not limited to performing multiple in-situ calibration scans using the CT system.

[0086] (6) A method according to any one of (1) to (5), comprising, without limitation, determining whether the plurality of calibration scans indicate that the performance of at least one pixel of at least one detector of the CT system is substandard based on the first reference set, reconstructing an image using the first reference set, and determining whether the reconstructed image includes a ring formed by at least one pixel of at least one detector of the CT system that is substandard.

[0087] (7) A method of any one of (1) to (6) which includes, but is not limited to, storing the reading of at least one pixel of at least one detector of the CT system that is determined to be of poor performance in a corresponding one of a defective spectral pixel table and a defective counting pixel table.

[0088] (8)(7) The method of saving the indices of at least one pixel of at least one detector of the CT system that has been determined to be performing poorly includes saving in a poor-counting pixel table that at least one pixel of at least one detector of the CT system that has been determined to be performing poorly has experienced at least one arbitrary combination of: (1) excessive background counting rate, (2) excessively high counting rate in a given X-ray beam, (3) excessively low counting rate in a given X-ray beam, (4) a ratio of count variation to the mean of counts greater than a first threshold, and (5) a deviation from the mean of counts greater than a second threshold.

[0089] (9) A method of any one of (1) to (8) which includes saving the indices of at least one pixel of at least one detector of the CT system that has been determined to be of poor performance, but also including saving a protocol for determining that at least one pixel of at least one detector of the CT system is of poor performance.

[0090] (10) A method of any one of (1) to (9) which includes performing multiple in-situ calibration scans using the CT system, but which includes performing in-situ calibration scans periodically to determine discrepancies in detector performance in the time domain.

[0091] (11) A method which further includes, but is not limited to, one of the methods of (1) to (10) performing a count-based scan of a subject using the CT system without using at least one pixel of the at least one detector.

[0092] (12) A method which further includes, but is not limited to, one of the methods of (1) to (10), performing a spectral-based scan of a subject using the CT system without using at least one pixel of the at least one detector.

[0093] The method of (13)(9) further includes, but is not limited to, scanning a subject using the CT system without using at least one pixel of at least one detector of the CT system for a protocol for determining that at least one pixel of at least one detector of the CT system is of inferior performance, and scanning a subject using the CT system with at least one pixel of at least one detector for a protocol other than the protocol for determining that at least one pixel of at least one detector of the CT system is of inferior performance. That is, the pixel determination method according to the embodiment scans a subject without using the inferior pixel for a protocol for determining the performance of an inferior pixel, and scans the subject with the inferior pixel for a protocol other than the protocol.

[0094] (14) A method which further includes scanning a subject using the CT system by adding values ​​from a macroblock containing at least one pixel of the at least one detector, without using at least one pixel of the at least one detector as part of the result of the addition, any one of the methods of (1) to (13) is further, but is not limited to.

[0095] A method according to (15)(7) wherein saving the indices of at least one pixel of at least one detector of the CT system that is determined to be of poor performance includes saving in a poor spectral pixel table that at least one pixel of at least one detector of the CT system that is determined to be of poor performance has experienced at least one arbitrary combination of (1) a corresponding normalized bin count above a maximum threshold and (2) a corresponding normalized bin count below a minimum threshold.

[0096] (16) A computed tomography (CT) system comprising a processing circuit configured to perform any one of the methods (1) to (15), without limitation.

[0097] (17) A computer system including a computer-readable medium for storing computer-executable instructions, which controls a computer processor under the control of the stored computer-executable instructions to perform any one of the methods (1) to (15).

[0098] Many modifications and changes can be made to some of the embodiments presented herein based on the above findings. Therefore, it is clear that the disclosure can be implemented within the scope of some of the claims described herein by means other than those specifically described herein.

[0099] According to at least one embodiment described above, it is possible to determine which pixels are inferior.

[0100] While several embodiments have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be implemented in a variety of other forms, and various omissions, substitutions, modifications, and combinations of embodiments are possible without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0101] 1156 Preprocessing device 1160 System Controller 1162 Storage device 1164 Reconfiguration device 1165 Input device 1166 Display device

Claims

1. Based on counting images and spectral images generated from multiple scans, the underperforming pixels of the detector are identified. The information of the determined pixel is stored, The inferior pixels are determined from the counted image according to the first judgment criterion. A second determination criterion, different from the first determination criterion, for the spectral image, A pixel determination method for determining which pixels are inferior based on a determination criterion that is based on the ratio of the variation in the count to the average value of the count.

2. The pixel determination method according to claim 1, wherein the plurality of scans perform at least one of (a) a plurality of air scans at different intensities, (b) a plurality of scans at different attenuation path lengths, and (c) a plurality of repeatability and / or reliability scans.

3. The pixel determination method according to claim 2, further comprising performing either a noise scan or a dark scan that does not use X-rays to measure the counting background.

4. The inferior pixels are determined by determining whether the multiple scans match the calibrated forward model. The pixel determination method according to claim 1.

5. The aforementioned multiple scans are scans using a calibration phantom. The pixel determination method according to claim 1.

6. A pixel determination method according to claim 1, comprising: reconstructing an image based on a first determination criterion; and determining the inferior pixels based on whether or not the reconstructed image includes a ring formed by the inferior pixels.

7. Based on counting images and spectral images generated from multiple scans, the underperforming pixels of the detector are identified. The information of the determined pixel is stored, The information of the inferior pixels is stored in the defective spectral pixel table or the defective counting pixel table. A pixel determination method that stores in a defective spectral pixel table that the aforementioned inferior pixel has experienced at least one of the following: (1) the normalized bin count exceeds the maximum threshold, or (2) the normalized bin count falls below the minimum threshold.

8. The aforementioned inferior pixels, The pixel determination method according to claim 1, wherein the pixel has experienced at least one of the following in a defective pixel table: (1) an excessive background count rate, (2) an excessively high count rate in a certain X-ray beam, (3) an excessively low count rate in a certain X-ray beam, (4) a ratio of count variation to the mean count that is greater than a first threshold, and (5) a deviation from the mean count that is greater than a second threshold.

9. Based on counting images and spectral images generated from multiple scans, the underperforming pixels of the detector are identified. The information of the determined pixel is stored, A protocol for determining the performance of the aforementioned inferior pixels is stored. For the protocol for determining the performance of the inferior pixels, the subject is scanned without using the inferior pixels, A pixel determination method for scanning a subject using the inferior pixels for a protocol other than the aforementioned protocol.

10. The pixel determination method according to claim 1, wherein the plurality of scans are performed periodically to determine the discrepancy in detector performance in the time domain.

11. The pixel determination method according to claim 1, wherein a count-based scan of a subject is performed without using the inferior pixels.

12. The pixel determination method according to claim 1, wherein a spectral-based scan of a subject is performed without using the inferior pixels.

13. Based on counting images and spectral images generated from multiple scans, the underperforming pixels of the detector are identified. The information of the determined pixel is stored, A pixel determination method that scans a subject while adding values ​​from blocks containing the inferior pixels, without using the inferior pixels as part of the summation result.

14. A control unit that performs multiple scans, A determination unit that determines which pixels of the detector have poor performance based on the counting image and spectral image generated based on the multiple scans, A storage unit that stores the information of the determined pixels, Equipped with, The determination unit determines the inferior pixels in the counted image according to a first determination criterion, and determines the inferior pixels in the spectral image according to a second determination criterion different from the first determination criterion, which is a determination criterion based on the ratio of count variation to the average value of the count, in an X-ray CT apparatus.

15. A control unit that performs multiple scans, A determination unit that determines which pixels of the detector have poor performance based on the counting image and spectral image generated based on the multiple scans, A storage unit that stores the information of the determined pixels, Equipped with, The memory unit stores information about the inferior pixels in the defective spectral pixel table. An X-ray CT apparatus that stores in a defective spectral pixel table that the aforementioned inferior pixel has experienced at least one of the following: (1) the normalized bin count exceeds the maximum threshold, or (2) the normalized bin count falls below the minimum threshold.

16. A control unit that performs multiple scans, A determination unit that determines which pixels of the detector have poor performance based on the counting image and spectral image generated based on the multiple scans, A storage unit that stores the information of the determined pixels, Equipped with, The memory unit stores a protocol for determining the performance of the inferior pixels. The control unit performs a scan of the subject without using the inferior pixels for the protocol for determining the performance of the inferior pixels, An X-ray CT scanner that scans the subject using the inferior pixels for protocols other than the aforementioned protocol.

17. A control unit that performs multiple scans, A determination unit that determines which pixels of the detector have poor performance based on the counting image and spectral image generated based on the multiple scans, A storage unit that stores the information of the determined pixels, Equipped with, The storage unit stores the information of the determined pixel, The control unit performs a scan of a subject while adding values ​​from blocks containing the inferior pixels, without using the inferior pixels as part of the summation result.

18. Based on multiple scans, pixels that are inferior are determined according to a first criterion for the count image generated, and based on the multiple scans, pixels that are inferior are determined according to a second criterion different from the first criterion, which is a criterion based on the ratio of count variation to the average value of the count. A program that causes a computer to perform a process to store the information of the determined pixels.

19. Based on counting images and spectral images generated from multiple scans, the underperforming pixels of the detector are identified. The computer is instructed to store information about the inferior pixels in the inferior spectral pixel table by saving to the inferior spectral pixel table that the inferior pixel has experienced at least one of the following: (1) the normalized bin count exceeds the maximum threshold, or (2) the normalized bin count falls below the minimum threshold. program.

Images