阅读说明：本技术 电荷共享校准方法及系统 (Charge sharing calibration method and system ) 是由 R·斯特德曼布克 E·勒斯尔 于 2019-01-08 设计创作，主要内容包括：信号处理系统(SPS)和相关方法。所述系统包括：输入接口(IN),其用于接收至少两个数据集,所述至少两个数据集包括第一数据集和第二数据集,所述第一数据集是由X射线探测器子系统(XDS)以第一像素大小生成的,并且所述第二数据集是以不同于所述第一像素大小的第二像素大小生成的；估计器(EST),其被配置为基于所述两个数据集来计算电荷共享影响的估计结果。(Signal Processing Systems (SPS) and related methods. The system comprises: an input Interface (IN) for receiving at least two data sets, the at least two data sets comprising a first data set and a second data set, the first data set being generated by an X-ray detector subsystem (XDS) at a first pixel size and the second data set being generated at a second pixel size different from the first pixel size; an Estimator (EST) configured to calculate an estimate of charge sharing impact based on the two data sets.)

1. An imaging apparatus (100), comprising:

an X-ray imaging apparatus (XI) comprising a detector subsystem (XDS) having an native pixel size (d);

a Signal Processing System (SPS), comprising:

a Combiner (COMB) configured to combine signals generated at the detector subsystem (XDS) in response to X-radiation exposure to obtain at least a first data set or a second data set, wherein one of a first pixel size and a second pixel size is the native pixel size, and wherein at least one of the first pixel size and the second pixel size is a multiple of the native pixel size (d);

an input Interface (IN) for receiving at least two data sets, the at least two data sets comprising the first data set and the second data set, the first data set being generated by the X-ray detector subsystem (XDS) at the first pixel size and the second data set being generated at the second pixel size different from the first pixel size; and

an Estimator (EST) configured to calculate an estimation of the charge sharing impact by means of a functional model comprising a correction factor (C) based on the two data sets.

2. Imaging apparatus as defined in claim 1, wherein the Combiner (COMB) comprises a binning circuit (S1) for binning the signals generated at the X-ray detector subsystem (XDS).

3. The imaging apparatus as claimed in any one of the preceding claims, wherein the Estimator (EST) is configured to obtain the estimation by forming the correction factor(s) based on forming one or more ratios from values of the first and second data sets.

4. Imaging apparatus as claimed in any of the preceding claims, comprising a Corrector (CORR) configured to charge-share correct a third data set generated by the detector subsystem (XDS) or another detector based on the estimation result.

5. Imaging apparatus of any one of the preceding claims, wherein the detector subsystem (XDS) is of energy-resolving type.

6. A signal processing method, comprising:

combining signals generated by a detector subsystem (XDS) having a native pixel size (d) in response to X-radiation exposure to obtain at least a first data set or a second data set, wherein one of a first pixel size and a second pixel size is the native pixel size, and wherein at least one of the first pixel size and the second pixel size is a multiple of the native pixel size (d);

receiving (S610) at least two data sets, the at least two data sets comprising the first data set and the second data set, the first data set being generated by the X-ray detector subsystem (XDS) at the first pixel size and the second data set being generated at the second pixel size different from the first pixel size; and is

Based on the two data sets, an estimation of the charge sharing impact is calculated (S620) by means of a functional model comprising a correction factor (C).

7. The method of claim 6, comprising:

correcting (S630) a third data set generated by the detector subsystem (XDS) or another detector subsystem for charge sharing effects based on the estimation result.

8. A computer program element, which, when being executed by at least one Processing Unit (PU), is adapted to cause the Processing Unit (PU) to carry out the method according to claim 9 or 10.

9. A computer readable medium having stored thereon the program element of claim 8.

Technical Field

The invention relates to a signal processing system, a signal processing method, an imaging apparatus, a computer program element and a computer readable medium.

Background

Some X-ray imaging devices (e.g., Computed Tomography (CT) scanners, radiographic devices, or other devices) use energy discriminating detector instruments. Unlike conventional detectors, which can only perform energy integration, energy discriminating detector systems are capable of analyzing the energy spectrum of X-radiation. Extraction of such additional information allows, for example, spectral imaging to understand the material composition of the imaged sample.

One type of such energy discriminating detector system is a direct conversion photon counting detector. They use a large number of unstructured semiconductors to convert X-radiation into detector signals. Structuring or "pixelation" is achieved by arranging a plurality of electrodes on a semiconductor. The electrodes record photon events caused by charge clouds formed in the semiconductor by impinging photons. The electrodes provide detector signals in the form of electrical pulses, which can be processed into spectral image data.

The undesirable phenomenon of "charge sharing" may occur in this type of detector or similar event counter. "charge sharing" is an effect in which more than one electrode records very identical photon events, which may interfere with the energy discrimination capability of the imaging device.

One way to reduce the effect of charge sharing is to use an algorithm that analyzes the signals detected by the different pixels. In the case of a charge sharing event, many pulses of smaller pulse height are detected in adjacent pixels at the same time. These pulse heights can be combined to restore the original pulse height.

Disclosure of Invention

Alternative methods may be needed to improve event count based imaging.

The object of the invention is solved by the subject matter of the independent claims, wherein further embodiments are incorporated in the dependent claims. It should be noted that the aspects described below are equally applicable to the imaging module and the imaging apparatus.

According to a first aspect of the present invention, there is provided a signal processing system comprising:

an input interface for receiving at least two data sets, the at least two data sets including a first data set and a second data set, the first data set generated by an X-ray detector subsystem at a first pixel size and the second data set generated at a second pixel size different from the first pixel size;

an estimator configured to calculate an estimate of charge sharing impact based on the two data sets.

In other words, the pixel size may be changed, and the data set represents the measurement result for each of the at least two pixel sizes. These measurements are then processed to assess the charge sharing impact in a given detector subsystem.

More particularly, and in accordance with one embodiment, the detector subsystem has a native pixel size, and wherein one of a first pixel size and a second pixel size is the native pixel size, and/or wherein at least one of the first pixel size and the second pixel size is a multiple of the native pixel size.

The pixel size (or area), sometimes expressed in terms of pitch, represents the effective area through which measurements can be collected and combined together (e.g., from a set of native pixels). In particular, the set of detector pixels is combined or binned into a single readout channel.

More particularly, and in one embodiment, the data set is obtained by operation of a combiner (which may be part of the system). The combiner is configured to combine signals generated at the detector sub-systems in response to X-radiation exposure to obtain at least a first data set or a second data set.

According to an embodiment, the combiner comprises a binning circuit for binning the signals generated at the X-ray detector subsystem. Other summing circuits may also be used instead or in combination.

According to an embodiment, the estimator is configured to form one or more ratios based on values from the first data set and the second data set, thereby obtaining the estimation result.

According to one embodiment, the system comprises a corrector configured to perform a charge sharing correction on a third data set generated by the detector subsystem or another detector based on the estimation result. The third data set is collected during an imaging phase, wherein the signals in the third data set are collected in respect of the object to be imaged, as opposed to early collecting the first data set and the second data set together as calibration data in a calibration phase preceding the imaging phase.

According to one embodiment, the detector subsystem is of an energy-resolving type (or photon-counting type).

According to a second aspect of the present invention, there is provided an image forming apparatus comprising: i) the signal processing system according to any one of the above embodiments; and ii) an X-ray imaging apparatus having the detector subsystem. The X-ray imaging apparatus may be a rotational, e.g. a CT scanner or a C-arm system, but other systems such as projection radiography systems are not excluded herein.

According to a third aspect of the present invention, there is provided a signal processing method comprising:

receiving at least two data sets, the at least two data sets including a first data set and a second data set, the first data set generated by the X-ray detector subsystem at a first pixel size and the second data set generated at a second pixel size different from the first pixel size;

an estimate of the charge sharing impact is calculated based on the two data sets.

According to one embodiment, the method comprises:

correcting a third data set generated by the detector subsystem or another detector subsystem for charge sharing effects based on the estimation.

According to a fourth aspect of the present invention, a computer program element is provided, which, when being executed by at least one (data) processing unit, is adapted to cause the processing unit to carry out the method according to any one of the mentioned aspects or embodiments.

According to a fifth aspect of the invention, a computer readable medium is provided, having a program element stored thereon.

The proposed system and method allow to support imaging with very small pixel size, in particular for use in energy-resolving photon counting detectors which are required to cope with very high X-ray flux. However, reducing the pixel size is constrained by the negative effects of charge sharing, which compromises energy performance. Thus, the pixel size is chosen to balance the energy resolution requirements of the flux capability.

The proposed system provides high rate capability at high throughput.

The proposed charge sharing compensation system may be implemented in software configured to process calibration data (the first and second data sets described above) obtained in different detector pixel configurations. The readout electronics may be adjusted by integrating the combiner described above to allow calibration data to be acquired at two or more effective pixel sizes. Calibration data acquired at large equivalent pixel sizes/pitches can be used to estimate the effect of charge sharing in separate calibration data sets acquired at smaller pixel pitches. This analysis of the calibration data allows the calibration data to be formulated to compensate for charge sharing on projection/image data acquired using nominally small pixel pitches that may be required to service all clinically relevant protocols.

Drawings

Exemplary embodiments of the invention will now be described with reference to the following drawings (not to scale), in which:

FIG. 1 shows a schematic block diagram of an X-ray imaging apparatus;

FIG. 2 shows a cross-sectional view of an X-ray detector module having an X-radiation component;

fig. 3 shows a schematic circuit diagram of an X-ray data acquisition system according to a first embodiment;

FIG. 4 shows a schematic block diagram of a signal processing system for charge sharing estimation;

fig. 5 shows a schematic circuit diagram of an X-ray data acquisition system according to a second embodiment;

FIG. 6 shows a flow chart of a signal processing method; and is

Fig. 7 shows a flow chart for the method steps of obtaining calibration data.

Detailed Description

Referring to fig. 1, a schematic block diagram of an X-ray imaging apparatus 100 comprising an X-ray imaging apparatus XI (also referred to herein as an "imager"), such as a computed tomography apparatus, projection radiography, or the like, is shown.

The imager XI is configured to generate images, in particular images relating to the internal structure and/or material composition of the object OB. The object OB may be animate or inanimate. In particular, the subject is a human or animal patient or part thereof.

Preferably, the X-ray imager XI is conceived to be used in particular for spectral imaging or photon counting (energy-resolved) imaging in the medical field, but other applications in non-medical fields, such as baggage scanning or non-destructive material testing (NDT) etc., are not excluded herein.

The X-ray imaging apparatus XI comprises an X-ray source XS configured to emit X-radiation.

An X-ray sensitive detector module XD is arranged at a distance from the X-ray source XS throughout the examination region. The X-ray sensitive detector module XD is coupled to a data acquisition circuit DAS. The X-ray sensitive detector module XD and the data acquisition circuitry DAS together form an X-ray detector (sub-) system XDs of the imaging apparatus 100. The module and DAS may be integrated in one unit or may be arranged to be discrete and separate but communicatively coupled.

The detector module XD is a transducer that converts X-radiation into electrical signals, which are then processed to digital (detector signals) by a data acquisition circuit DAS (referred to herein simply as "DAS") (in a manner described in more detail below). The detector signals may be processed by an image processor IP into the desired image. Depending on the number or contrast of images desired by the person, the image processor implements suitable image processing algorithms, such as filtered back-projection (for 3D imagery), phase contrast reconstruction, dark-field reconstruction, transmission reconstruction, or any combination thereof. The imagery may be rendered for display on a display device DU (e.g., a monitor) or may be otherwise further processed or may be stored in a memory DB.

During imaging, an object OB to be imaged (or a part thereof) is located in an examination region between an X-ray source XS and an X-ray detector XD. A user energizes X-ray source XS via a control unit (not shown). X-ray source XS then emits X-radiation in the form of X-ray beam XB, which passes through the examination region and object Ob to be imaged. The X-ray beam is composed of photons of different energies, which are defined by the spectrum of X-radiation generated by X-ray source XS.

The X-photons interact with the substance in the object OB. For example, some photons are absorbed by the substance, while other photons appear on the far side of the object (viewed from the X-ray source) and then interact with the X-ray sensitive detector XD. Some of the photons present distally of the object OB have been scattered because they have interacted with the substance in the object OB, while other photons have not been scattered. An anti-scatter grid (not shown) may be used to prevent scattered photons from reaching the detector system, thereby improving image quality.

Each photon in the X-ray beam has a certain energy. As primarily contemplated herein, the X-ray imager XI is capable of event counting to quantify the manner in which photons interact with the detector module XD. In a particular embodiment, the X-ray imager is a spectral imager, which allows spectral analysis of the detected X-radiation/photons. This capability allows, for example, material decomposition of the object. That is, for example, the detected radiation can be analyzed to identify different types of material tissue in the subject.

More particularly, the photons, after interaction with the object OB, interact with the X-ray sensitive layer of the X-ray detector XD, causing electrical signals which are then picked up and processed within the DAS.

The imaging apparatus 100 as proposed herein comprises a novel signal processing system SPS configured to correct the detector signal for charge sharing effects occurring in the detector system XDS. Charge sharing effects can corrupt the fidelity of the detector signal. If left unaccounted for, charge sharing can undermine image quality and the degree of spectral separation.

Before turning to the operation of the newly proposed signal processing system SPS, reference is first made to fig. 2 to illustrate the charge sharing effect in more detail. Figure 2 shows a cross-sectional view through the detector subsystem XDS, in particular part of the X-ray module XD and DAS. Which is parallel to the main propagation direction p of the X-radiation XB (see fig. 1).

The detector module XD is preferably of the direct converter type. More particularly, the detector XD comprises a direct conversion layer DL as an X-ray radiation sensitive component. The direct conversion layer DL is formed of a suitable semiconductor. The semiconductor has a crystal structure, for example, silicon, CdTe, CZT, GaAs, Ge, and the like. Like the entire detector XD unit, the conversion layer DL is typically rectangular and forms an array. In the view of fig. 2, the other length dimension of the layer extends into the plane of the paper in the figure. The direct conversion layer DL functions as a transducer. In other words, impinging photons in and through the layer DL generate an electrical signal. In particular, the conversion layer is sandwiched between a pair of electrodes EL. Only the anode arranged at the distal face of the conversion layer DL is shown in the cross-sectional view of fig. 2. The electrodes EL are discretely spaced at a distance D and in a pattern on the distal surface of the direct conversion layer DL. Each electrode EL has an original size (effective area) d. The size may also be expressed as a pitch. The electrode EL defines the "pixelation" of the originally unstructured conversion layer DL. Each electrode EL corresponds to a detector pixel, of which only three pixels PX1-3 are exemplarily shown from the entire pixel array. The electrode EL is sometimes referred to herein simply as a "pixel". The electrode EL may be arranged as a layer of a TFT (thin field transistor). A voltage is applied across the electrodes and the direct conversion layer DL. Typically, the cathode is not pixelated to apply the same voltage across the layer DL. X-ray photon Xph impinges on the crystals in detector layer DL. Depending on the energy of the photon Xph, some of the electrons and holes that were originally incorporated in the crystal are released. The electrons and holes thus released may themselves release more electrons and holes, and so on. Due to the applied voltage, a major portion of the electrons and holes cannot recombine and form a corresponding charge cloud CC. Driven by the applied voltage, the electron charge cloud CC drifts towards the anode EL (downwards in the view of fig. 2) to cause the earlier mentioned electrical signal (in particular the electrical pulse). The electrical signals are then processed by the DAS.

Each count of photon energy represents an event. The event counting capability of an imager and its fidelity depend to a large extent on its ability to distinguish between charge clouds caused by different photons. Ideally, therefore, each electrode pair EL will respond to a charge cloud of a single photon at a time. Unfortunately, this does not always occur because the finite size of the charge cloud is not negligible. If a cloud of charges is generated between adjacent pixels EL (as shown in fig. 2), a part of the cloud can be directed to one of the pixels by a corresponding electric field, while another part is directed to the other pixel(s). In other words, the subsequent cloud charge CC may be recorded by two or more (in particular adjacent) electrodes. This adverse effect is referred to as "charge sharing". In other words, the cloud charge CC caused by the single photon Xph through interaction with the direct conversion layer is shared between two or more anodes/pixels. This charge sharing can result in double or multiple counts for a single given photon, as charge is shared between two or more of the pixels defined by the anode. Charge sharing can therefore interfere with the energy discrimination capability of the imager XI and ultimately destroy the fidelity of the image.

It should be appreciated that the above-described architecture of the detector module XD is merely exemplary, and is intended to illustrate the signal chain from the X-ray photons Xph to the electrical signals in the case of charge sharing. Many other modifications to the above-described design in fig. 2 are also contemplated herein. In particular, in an alternative, a detector module XD of the indirect conversion type is also envisaged, which comprises a scintillator layer (instead of the layer DL) and a photodiode layer coupled thereto. In the case of the indirect conversion type, the pixelation is caused at least in part by crystal growth or mechanical structures in the scintillator.

Referring now to fig. 3, fig. 3 shows more details of a readout circuit of a DAS as contemplated herein according to one embodiment. The left part shows an exemplary detector pixel. For clarity, only two pixels PX1, PX2 are shown.

The pixels PX1, PX2 generate electrical (current) pulses having an amplitude ("height") whose area largely corresponds to the energy of the impinging photon Xph. The height of the electrical pulse detected at a given pixel PX1, PX2 is a function of the energy of the impinging photon Xph. The higher the photon energy, the higher the pulse amplitude that can be detected at the corresponding pixel PX1, PX 2.

Each pixel electrode PX1, PX2 is coupled to the photon counting circuitry of the DAS via a separate raw signal line (or "(pixel) readout channel") CH.

According to one embodiment, the electrical pulses generated at the pixels PX1, PX2 are processed by the photon counting circuitry in the following manner: the optional conditioning circuitry includes a preamplifier 320, which preamplifier 320 amplifies each electrical signal generated by any of the pixels 218.

The optional conditioning circuit may further include a pulse shaper (not shown) to process the amplified electrical signal for the detected photon and generate a corresponding analog signal including a pulse height (e.g., voltage/current or other pulse indicative of the detected photon). The pulses so generated have a predefined shape or profile. In this example, the pulse has a peak amplitude indicative of the energy of the detected photon.

The energy discriminator 325 energy discriminates the analog pulses. In this example, the energy discriminator 325 includes a plurality of comparators 330, each of the plurality of comparators 330 comparing the amplitude of the analog signal to a respective threshold corresponding to a particular energy level. The adjacent thresholds define energy bins. In other words, discriminator 325 operates to determine the "height" of the input pulse generated by the shaper. More particularly, each comparator 330 generates an output count signal that indicates whether the amplitude of the pulse exceeds its threshold. In this example, the output signal from each comparator produces a digital signal that includes a low-to-high (or high-to-low) transition when the pulse amplitude increases and crosses its threshold, and a high-to-low (or low-to-high) transition when the pulse amplitude decreases and crosses its threshold.

In an exemplary comparator embodiment, the output of each comparator transitions from low to high as the amplitude increases and crosses its threshold; and the output of each comparator transitions from high to low when the pulse amplitude decreases and crosses its threshold.

Counter 335 counts rising edges (or falling edges in some embodiments) for each threshold, respectively. For each threshold, counter 335 may comprise a single counter or separate sub-counters. Optionally, in case of only two-sided binning, there is an energy bin sorter (not shown) that sorts the count energy into bins that are in an energy range or correspond to a range between energy thresholds. Indeed, in a preferred embodiment with high throughput, there is no binning to range operation, but only the count of threshold crossings are recorded (i.e., one-sided binning).

The count data (denoted herein as M, which will be described in further detail below) may then be used to energy-resolve the detected photons. In other words, the photon counting circuitry of the DAS is operative to quantify the pulse height of each incoming pulse from each pixel PX1, PX2 into an energy bin defined by the number of voltage thresholds. K (K ≧ 2) (voltage, amperage, or other physical quantity indicative of energy) thresholds can define K different energy bins for recording pulse heights above the respective threshold of the thresholds. For example, a pulse whose edge rises above (i.e., "crosses") two of the thresholds will cause a count for each of the two bins associated with the respective two thresholds. If only the lower one of the thresholds is crossed, there will be only one count, etc. This is merely an example, as in some embodiments only falling edges cause counting or both rising and falling edges cause counting.

The photon counting circuit provides at its output for each pixel PX1, PX2 the number of counts per bin recorded per unit time. These photon count rates for each bin and pixel form projected photon count data, which can formally be written as M ═ M (M)₁,…,m_k)ⁱWherein the vector of the counting rate is m_kAnd i denotes the corresponding pixel, and 1. ltoreq. k.ltoreq.N is the number of energy bins used. In other words, m_kRepresenting the number (count) of pulses per unit time that have been recorded at pixel i whose height falls into bin k. The counts may be normalized by the frame rate to represent the count rate, i.e., the count per unit time. However, normalization is not necessary, and the proposed system can also operate on count data that is not normalized. There are 2, 3 or more energy thresholds. In a rotating system such as a CT or C-arm, the recorded count rates may be different for different projection directions, so the above concept can be supplemented with an additional index for the projection directions. In the latter case, M forms a sinogram in the CT embodiment.

The event count data M thus quantized may then be processed by an image processor IP.

Referring now to fig. 4, fig. 4 shows a schematic block diagram of a signal processing system SPS configured to perform charge-sharing correction of data in or generated by the detector subsystem XDS during operation of the imager XI.

The X-ray detector subsystem XDS is configured to be operable in a plurality of detector configurations. The data generated by the subsystem XDS at different detector configurations under calibration exposures can then be compared to assess or quantify the charge sharing effects. The detector configuration is defined by the pixel size PX of the detector unit XD. One such size (i.e. the original size d) of the detector pixels is given by the hardware means that produce the pixelation, as illustrated in fig. 2. For the case of a direct conversion detector, the pixel size D may be given by the size of the electrodes in the readout layer, as explained above in fig. 2. Alternatively, the original size may be given by the crystal growth in the scintillation layer and/or the size of the photon detector used in case of indirect conversion. In short, native pixel size is the smallest possible size at which the imager can operate to provide the best spatial discrimination. The native pixel size d is the smallest physical arrangement in the detector unit that converts X-rays into electrical pulses.

Pixel sizes for different detector configurations may be achieved by combining (e.g. binning) the detector signals for a plurality of such native pixels to achieve different detector configurations and thus different (virtual or effective) pixel sizes. For example, the data signals for four native pixels may be combined to thus achieve a "virtual" detector pixel having four times the native pixel size d. Such a configuration may be referred to as "P to 1," i.e., one virtual pixel has P native pixels, where P is a natural number greater than 1. In particular, the actual value of P comprises 2, 3, 4 or more, such as 8, or even up to two digits. For clarity, P is used as 1, denoted in this notation as "1 to 1", which is an arrangement corresponding to the usual native pixel size P arrangement. Different detector configurations can be realized by means of a signal combiner function COMB.

Referring again to the circuitry in fig. 3, fig. 3 shows such a combiner function COMB according to one embodiment, which is integrated into the X-ray subsystem XDS, in particular into the photon counting circuitry of the X-ray subsystem XDS. In particular, the combiner COMB is implemented as a binning, which bins the two pixels PX1 and PX2 into a single channel to generate a 2-to-1 detector configuration, thereby reducing the spatial discrimination by half. To this end, the combiner COMB comprises a switch S1,As shown in fig. 3. The switches are arranged to pick up and combine electrical pulses travelling along the readout channel CH, where "-" represents a negative operator with respect to both states "open (off)", "closed (on)" of the switch S1. S1 is arranged in two readout lines for PX1, PX2On the cross-link line XL in between. In particular, switch S1, when switched from off to on, will cause switch S2 (e.g., disposed on the readout line of PX 2) to disconnect pixel P2 from its native pitch readout channel CH and connect pixel P2 to the input node corresponding to the readout electronics of pixel P1. The pulse signals generated at either of the pixels PX1, PX2 will add by superposition and are now located on a single readout channel (e.g., the readout channel of pixel PX 1) and processed by the photon counting circuitry of pixel PX 1. It should be understood that the charge sharing events across PX1 and PX2 (simultaneous in time) will be added and considered as a single event at the correct deposition energy at the thresholder 325 and discriminator 330 for pixel PX 1.

Fig. 5 shows an alternative combiner COMB arrangement for binning, wherein the combiner functions COMB are integrated into different stages of the photon counting circuit. In fig. 3 the combiner acts on the readout signal line CH to pick up and combine the pulses, whereas in fig. 5 the combiner COMB is instead arranged downstream and forms the combined signal at the output stage of the amplifier 320 or shaper. That is, the charge pulse signals themselves are not combined, but the outputs of the respective shapers are so combined. S1 is now arranged on the feed lines from the shaper/preamplifier 320 to the respective discriminators 325, while S2 is arranged on the respective cross line X1 between the feed lines.

Alternatively, the combiner function COMB may be integrated in other stages of the DAS, in particular in other stages of the photon counting circuit, for example at the pulse counter 335 itself. This combiner embodiment may then involve calculating the resulting equivalent energy, since the count may not be added as simply when combining, but rather the combining operation is performed by the system increasing the count by 1 in bins for higher thresholds based on the resulting equivalent energy.

The embodiment in fig. 3 is more advantageous than the embodiment in fig. 5 in terms of robustness to the sum of the squares of the noise contributions, because in fig. 3 the noise from all binned readout electronics is combined into one single node. Then, as used herein, the term "detector signal" may refer to any pulse or combination of pulses, regardless of where in the detector subsystem XDS the pulses are picked up (whether at the readout line CH (e.g., in fig. 3) or other downstream stages (e.g., in the embodiment of fig. 5)).

Although the particular embodiments of the combiner COMB in fig. 3, 5 are configured to produce a 2 to 1 configuration, it should be understood that the concept can be extended to several other configurations, e.g., 3 to 1, 4 to 1, 5 to 1, etc., for any P > 2. That is, the switch may be operated to add not only two pulses, but also more than two (e.g., 3, 4, or 5 or more) pulses and any sub-sums thereof. For example, a pulse sum consisting of 5 pulses from 5 different pixels can be formed for an arrangement of e.g. 5 to 1, and for any k pulses selected from these 5 pulses, preferably each sub-sum is less than 5 pulses. In other words, the switching network at the input nodes can be made and controlled by appropriate logic, enabling multiple detector configurations to be implemented, thereby allowing at least 2 different detector configurations to be obtained for any selected k (where 1 < k ≦ P), resulting in any of P different P to 1 configurations. One way to implement such a network is to add cross-link lines XL between some or all (or at least P) pixel readout lines, with a corresponding switch on each cross-link line and one switch on each pixel readout line CH (of the P pixel readout lines). The complexity of the network can be reduced by Karnaugh-vent graph analysis or other digital synthesis tools.

A single combiner COMB (comprising two switches S1, S2) as shown in fig. 3, 5 for the entire detector XD is sufficient, but preferably a plurality of such similar combiners with corresponding switches S1, S2, respectively, are arranged between a plurality of pixels, not only for the two pixels PX1, PX2 as shown. In particular and preferably, for any two adjacent pixels, there is a combiner COMB similar or analogous to the one example shown in fig. 3, 5. For example, another such combiner (not shown) is arranged between the pixels PX3 (not shown) and PX2 to combine the pulse of the short seed PX3 with the pulse from PX2, or the like.

Not all detector configurations may be ideal for each imaging setup, so the combiner may not be universal, but may be pre-configured to meet the specific requirements of the imager XI. For example, in a CT setting where the native pixel pitch D is 500 μm and the pitch of the anti-scatter grid is 1mm, the combiner can be limited to, for example, a 2-to-1 configuration and a 4-to-1 configuration within the ASG wall. To this end, the combiner may not need to be configurable for combinations beyond a set of native pixels confined within a region between two adjacent ASG walls. Thus, the combiner may be configured (e.g. hardwired) for a fixed and limited number of different (at least two) detector configurations. However, in other embodiments, user-configurable combiners that are configurable for all possible configurations up to and including a given P configuration are also contemplated herein.

It should be understood that with respect to the negative switch pair S1,The implementation of each combiner function COMB is only one of the other embodiments equally envisaged herein in the alternative. However, the switch pair embodiment is the preferred embodiment due to its simplicity, reliability. Switch types contemplated herein include, for example, transistors, such as metal oxide semiconductor field effect transistors (MOS-FETs), which have low series resistance when in an "on" state and low leakage when in an "off" state.

Referring again to fig. 4, the signal processing circuit SPS comprises a suitable interface unit CINF and an associated controller by which the SPS can request the X-ray detector system XDS to operate with at least two of the desired ones of the different detector configurations. In this manner, respective calibration data MDC1 and MDC2 (which are denoted herein as DC1 and DC2) can be generated for respective probe configurations. "DC 1" is used herein to denote the native pixel size detector configuration. As described above, the number of pixels N to be combined together is user configurable. Furthermore, it is not necessary to perform calibration for each pixel of the detector array XD, but it is sufficient to calibrate a single group of pixels (anywhere on the detector array XD) with a desired number of pixels (two or more, e.g., three or four or more). Measuring only a subset of pixels allows a reduction in the amount of data to be processed.

The respective calibration data as well as the MDC1 and MDC2 are received at the input port IN of the signal processing system SPS after the X-ray detection system XDS has been operated IN two or more detector configurations.

The estimator EST then estimates or quantifies the charge sharing using the (at least) two data sets MDC1 and MDC 2. In one embodiment, the estimator EST uses a functional model to estimate the charge sharing influence in the form of the correction value C. These correction values are specifically associated with the smaller pixel size detector configuration DC1, as this is expected to be more affected by charge sharing.

One way to estimate the charge sharing impact fs is by forming a ratio from the detector signals generated separately for the two detector configurations. For example, for a P to 1 configuration (P is a natural number ≧ 2), the ratio can be found according to the following equation:

c ═ P × MDC1/MDC2 or MDC2/P × MDC1 (1)

Where MDC1 is the detector signal configured for the original size, and MDC2 is the combined signal for the configuration with P multiplied by the original size P. For example, if detector signals relating to P pixels (e.g., P ═ 4) are combined, the value will be different due to charge sharing than the value resulting from multiplying P by the detector signal in the smaller pixel configuration DC 1. Therefore, the ratio C will not be equal to 1. Thus, when the detector operates as usual (without a combiner) with a detector configuration DC1 of smaller pixel size, the ratio C as described above can be used as a correction factor to be applied to the detector readings. For estimation purposes, forming functional expressions other than ratios, e.g., weights or absolute differences, are also contemplated herein, as will be explained in more detail below with respect to fig. 6. It should be noted that the ratio C is typically formed for each respective count of each bin of the counter 335. In other words, the ratio C is formed for each energy bin.

For the user or by obtaining two data sets MDC1, MDC2 by means of a random generator, a single pixel, e.g. PX1, is specified. This specification can be realized by selecting the coordinates of the pixel PX1, for example. Each pixel can be addressed by a unique coordinate. For example, the pixel pattern may be rectangular, wherein the pixels are arranged in rows and columns. Each pixel has unique coordinates (x, y), x being its row and y its column. Other addressing schemes may also be used. In addition, it is preferable to select a group of two or more pixels in the neighborhood around the pixel PX 1. For example, in the grid-like arrangement of the pixel array XD, the 4 neighboring pixels around the pixel PX1 are thus selected.

The detector and the group of pixels and pixel PX1 are then exposed to radiation by operating the X-ray source in a calibration phase. Two measurements are then made, one for each configuration, e.g., one for the raw size (DC1) and another for the larger pixel size configuration (i.e., P for 1 (e.g., P-4) DC 2). More than two exposures are required if more than two detector configurations are used. In the exposure for the DC2 configuration, the combiner COMB operates to combine the detector signals relating to the groups of pixels and the pixels PX1 to form two data sets MDC1, MDC 2. As a result, two data sets MDC1, MD2 are obtained in the calibration phase. The data set MDC1 comprises the (conventional) detector signal recorded by the counter 335 as a cross bin count for the pixel PX1, while the other data set MDC2 comprises a combined signal from P-4 pixels in the group, which combined signal is recorded as a count in the bin of the counter 335. Thus, in principle, each data set MDC2, MDC1 includes a single number for each energy bin, and these numbers can be used to estimate the charge sharing impact according to C in equation (1) above, where P is 4. Other P values, for example, 2, 3, 5 or greater, may alternatively be used.

This same correction factor C can be used for all other pixels during the actual imaging phase (after the calibration phase described above) to correct for charge sharing. Alternatively, and preferably, the above operations are repeated for a plurality of different pixel locations PXj (particularly for all pixels in the array XD), and then the individual correction factors Cj are averaged or otherwise arithmetically combined.

Alternatively, and still more preferably, the correction factors Cj may be retained pixel by pixel, dedicated to each respective pixel j, and the set of correction factors Cj is stored in the memory CEM, each correction factor Cj being associated with a respective pixel position j. The correction operation then includes a lookup operation to retrieve the associated correction factor Cj for each pixel PXj during or after imaging.

Preferably, and in order to obtain better results, the above-described procedure for obtaining calibration measurements MDC1, MDC2 is obtained in the case of energy calibration and/or material calibration (described further herein below at fig. 7) and/or in the case of different ranges of impinging X-ray flux. The data sets MDC1, MDC2 form a multidimensional collection of numbers, indexed by respective binning (i.e., energy), and indexed by at least material type/thickness. Additionally and optionally, indexing by flux range or other parameters may be used. As a result, correction data C is again a multidimensional array of numbers, the entries of which are associated by appropriate indices with the corresponding pixels, bins, materials, energies, etc., or other relevant factors.

Next, IN the imaging mode, after the above-described calibration mode, detector signals (with respect to the actual object OB to be imaged) from the X-ray detector subsystem XDS are then received at the input port IN. In the imaging mode, no calibration is required, and the detector signals of the object OB are then passed directly to the corrector CORR.

The corrector CORR then retrieves the associated correlation correction values Cj stored in the memory CEM and associates them with the detector configuration (for example DC1) used for the current imaging. The corrector then applies (e.g. multiplies) the correction value Cj to the received detector signal of the object OB and forms a corrected detector value, which is then output at the output port OUT.

The image processor IP can then use the thus corrected detector signals to calculate the desired imagery, for example, transmission images, phase contrast images, dark field images or spectral images or other images.

As can be appreciated from the above description, the charge sharing impact evaluation on the data generated by the subsystem XDS can be customized for a particular imaging apparatus XI, a particular detector cell XD, a particular pixel portion of the detector cell XD, even down to the individual pixel level. The proposed subsystem can be retrofitted to an existing imager XI.

The operation of the signal processing system SPS will now be described in more detail with reference to the flow diagrams of fig. 6, 7. It should be understood, however, that the method steps to be described below are not necessarily limited to the architectures according to fig. 1-5. In particular, the method steps described below can also be understood as constituting teaching per se.

Turning first to fig. 6, in a preliminary step S605, respective data sets ("calibration measurements" or calibration data) are acquired in a calibration run for respective different detector configurations. Each configuration is determined by the pixel size used and the pixel size is different for each set. Different calibration measurements M ═ (MDC1, MDC2) can be obtained by a combiner circuit COMB which combines (in particular sums) the detector signals associated with the different pixels PX1, PX 2. This combination may occur at any stage of the X-ray detector subsystem XDS. In particular, the combination may occur in the analog phase or when the signal is digitized by a counter stage. The operation of the combiner results in different counts generated by the photon counting circuitry (which form the data set M) according to two different configurations, it is these counts that are designated herein as calibration measurements ("calibration data") M ═ MDC1, MDC 2.

Preferably, one of the two or more detector configurations is set to the smallest possible case, i.e. the original pixel size, while the other detector configuration is selected to be a suitable P-times thereof, e.g. twice or four times or eight times the original configuration. In principle, however, as previously mentioned, P may be any number greater than 2. The native probe configuration is denoted herein as DC1, while MDC1 denotes the associated calibration measurements.

Then in step S610, two (or more) calibration sets MDC1 and MDC2 are received at the data processing unit PU (e.g. a computer unit of a workstation or a computing function integrated into the imager XI).

At step S620, an estimation result of the charge sharing influence is calculated. Such an estimation result or a quantization result of the influence on the charge sharing may be represented as correction data, for example, a correction value or a correction factor. In this estimation, the functional model f is used to combine the measurements for two (or more) different detector configurations DC1, DC 2.

In general, given a function model f, the correction data is:

C_b＝f(MDC1_b,MDC2_b) (2)

the index "b" distinguishes the counts MDC1 in the individual energy bins b for the (at least two) different detector configurations DC1, DC2_b、MDC2_b。

A functional model f is a linear dependence which yields the ratio mentioned above at (1), so in this case f is:

in another embodiment, an additive function model is used, such as:

C_b＝f(MDC1_b,MDC2_b)＝(m*MDC1_b-MDC2_b)^kwherein k is not less than 1 (4)

In (4), if there is no charge sharing, C will be reduced to zero and any deviation therefrom can be used as an indication of the amount of particular charge sharing. If k is 1, the difference can be taken as the absolute value of the signed sum. The square deviation k 2 can also be used advantageously.

Although reference has been made above primarily to the use of two probe configurations DC1, DC1 and associated calibration data M (MDC1, MDC2), the method can also be extended by taking three or more (p > 2) in a data set at three or more probe configurations and by compiling this information to extract correction data therefrom and an estimation result for charge sharing. For example, the functional model may include forming a more complex ratio or (weighted sum). That is, more than two (e.g., three, four, or more) detector configurations p can be functionally and arithmetically combined to quantify the charge sharing impact according to the following equation:

f(MDC1_b，MDC2_b，…，MDCp_b) (5)

for example, one can deal with a configuration for n (among others)Seed configuration) of j, k_j、MDC_kAnd a ratio is formed for any pair j, k according to the above formula. Then contrast ratioAveraging (weighted or mean) is performed to obtain the correction factor C.

As a further extension of the above, it may not always be necessary to configure one of the configurations as the native detector pixel size configuration. For example, one configuration may be formed to include groups of pixels greater than 1, while a second (third, etc.) configuration also involves groups of different sizes. In other words, if the situation requires downsampling of the imaging task, it is also contemplated to use a P-to-L configuration where L is greater than 1.

It has been found that the above quantization in terms of ratio with equations (1), (3) of the charge sharing effect is a good approximation to the low pass setting. However, in higher flux settings, the above-described linearization with equations (1), (3) can be refined by modulation with an exponential term, modeling the impact of high flux and the connection to the detector electronics, in particular dead time and/or abnormal operating capability. In particular, expression (3) may be refined as:

where k is a constant less than 1 (e.g., 3/4), v is the flux of the larger pixel DC2, and τ is the dead time of the detector system XDS.

The above-described relationship of any of the models in equations (1) - (6) can be further refined by including one or more other terms representing the pulse pile-up model.

In the previous estimation embodiments (equations (1) - (6)), it was assumed that the binning of two or more pixels has no influence on the transient response of the direct conversion material. This can be considered a good approximation for a medium binning configuration (e.g. 4 x 500 μm versus 1mm in a 4-to-1 configuration). For larger (e.g., 8-to-1) equivalent pixels, significant changes in transient response (e.g., longer transient response for large pixel configurations) may result due to different weighted potential distributions (large pixels may not benefit from the so-called small pixel effect). Such different transient responses may affect the signal generation of the preamplifier 320 or similar components of the front-end electronics in the DAS. One example of such an effect due to different transient responses is ballistic deficit. To this end, it may be necessary to use different energy calibrations for each detector configuration to achieve comparable results.

Then, at step S630, the estimation result (in particular, the correction data C) can be used to correct the detector data M' generated by the imager XI during the imaging operation of imaging the object OB for charge sharing. The imaging operation is preferably one in which the detector configuration employed corresponds to the small pixel size configuration DC1 used earlier in the calibration phase. In other words, the correction data C according to (1) and (3) is preferably associated with the configuration DC1, and therefore C may be used^DC1To write (1), (3) to conceptually better indicate this dependency. However, canTo use the reciprocal C of the correction data^-1To correct for data obtained when imager XI is operated at a larger pixel size under configuration DC 2.

Either the correction data are applied directly during imaging, which is preferred, or alternatively the (as yet) uncorrected detector readings M' relating to the imaged object OB are first stored or buffered and the correction data C are applied at a later stage, for example when visualization is required. Depending on which function model is used in step S620, correction can be achieved by multiplying the correction value by the image data M' (formula (1)) or subtracting or adding (formula (4)).

Although the above method may be performed by simply exposing the X-ray detector to radiation during an air scan, it is preferred to collect calibration data in the context of an energy calibration scheme and a material calibration scheme. Referring now to fig. 7, this is illustrated in more detail in the flow chart of fig. 7. In other words, the flowchart in fig. 7 provides more details on how to perform method step S605 of generating two or more calibration data sets.

At step S710, an energy calibration is performed such that all pixel thresholds in discriminator 325 are set to exactly the same energy across the entire detector array XD. In particular, the energy calibration allows for factors of gain and offset to be taken into account.

In this calibration state of the detector (sub) system, a material calibration is performed at step S710, where beam XB has a specific configuration of calibration material c therein. Suitable materials ("phantoms") are any of water, Delrin, tin, Teflon or k-edge materials (e.g., AU, Bi, Pb) or other materials. The material should have at least two different thicknesses. In this way, one exposure may be sufficient, or multiple items of the material may be otherwise stacked to achieve different thicknesses, and two additional exposures may need to be run. Material calibration allows for the configuration of a curve or look-up table (LUT) to convert count rates to material thickness for imaging. The more material thickness used, the more accurate the LUT. For a given count rate, the associated thickness may then be interpolated from the LUT.

Will use MDC1_c,bCorresponding calibration data for the first detector configuration DC1 (native detector configuration) is represented, where the index c distinguishes the calibration material (or thickness) c in the beam.

Similarly, after switching the probe configuration by performing binning as described above, the energy calibration and material calibration are repeated at step S730, resulting in a corresponding measurement MDC2_c,b. MDC1 due to measurement results_c,bAnd MDC2_c,bCorresponding to exactly the same irradiation conditions, so the assumed relationship f (as described above) that exists between them is influenced by the amount of charge sharing. Optionally, in a further step, measurements for a series of different flux settings may be obtained, wherein the correction C is further indexed by the flux rate.

The above-described flow in steps S710-S730 shows the above-described multidimensional nature in the most general case of correction data C, due to the various dependencies on b, C, pixel position (x, y), detector configuration DCj. However, in some cases, with appropriate simplification, the dependencies may be reduced to binned dependencies.

The components of the image processing system SPS may be implemented as software modules or routines in a single software suite and may run on a general purpose computing unit PU, such as a workstation associated with the imager XI or a server computer associated with a group of imagers. Alternatively, the components of the image processing system ISP may be arranged in a distributed architecture and/or "cloud" and may be connected in a suitable communication network.

As a further alternative, some or all of the components of the SPS may be arranged in hardware (e.g. a suitably programmed FPGA (field programmable gate array)) or may be arranged as a hardwired IC chip (e.g. an ASIC (application specific integrated circuit)) on a PCB module included in the circuitry for the detector subsystem XDS.

Although in the above the converter is converted into an effective material path length l, this should be broadly considered, as any other parameter that is converted into a length equivalent to said effective path length is also contemplated herein. Further, mathematically equivalently reformulated formulas relating to any of the above formulas are also contemplated herein.

In another exemplary embodiment of the invention, a computer program or a computer program element is provided, which is characterized by method steps adapted to run the method according to one of the preceding embodiments on a suitable system.

Thus, the computer program element may be stored in a computer unit, which may also be part of an embodiment of the present invention. The computing unit may be adapted to perform or cause the performance of the steps of the above-described method. Furthermore, the computing unit may be adapted to operate the components of the apparatus described above. The computing unit can be adapted to operate automatically and/or to run commands of a user. The computer program may be loaded into a working memory of a data processor. Accordingly, a data processor may be equipped to perform the methods of the present invention.

This exemplary embodiment of the invention covers both a computer program that uses the invention from the outset and a computer program that is updated by means of an existing program to a program that uses the invention.

Further, the computer program element may be able to provide all necessary steps to complete the flow of an exemplary embodiment of the method as described above.

According to a further exemplary embodiment of the present invention, a computer-readable medium, for example a CD-ROM, is proposed, wherein the computer-readable medium has a computer program element stored thereon, which computer program element is described by the preceding sections.

A computer program may be stored and/or distributed on a suitable medium, particularly but not necessarily a non-transitory medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the internet or other wired or wireless telecommunication systems.

However, the computer program may also be present on a network, such as the world wide web, and may be downloaded into the working memory of a data processor from such a network. According to a further exemplary embodiment of the present invention, a medium for making a computer program element available for downloading is provided, the computer program element being arranged to perform the method according to one of the previously described embodiments of the present invention.

It has to be noted that embodiments of the invention are described with reference to different subject matters. In particular, some embodiments are described with reference to method type claims whereas other embodiments are described with reference to apparatus type claims. However, unless otherwise indicated, a person skilled in the art will gather from the above and the following description that, in addition to any combination of features belonging to one type of subject-matter, also any combination between features relating to different subject-matters is considered to be disclosed with this application. However, all features can be combined to provide a synergistic effect more than a simple addition of features.

While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive. The invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

In the claims, the word "comprising" does not exclude other elements or steps, and the word "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. Although some measures are recited in mutually different dependent claims, this does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims shall not be construed as limiting the scope.