Method for correcting a spatially resolved photon measurement of an X-ray detector

Abstract

Method for correcting a spatially resolved photon measurement of an X-ray detector (2), wherein in the X-ray detector (2) a signal contribution (14a, 14b) in a first electrical signal (34) is generated with spatial resolution from an incident X-ray photon (4) by means of a current-voltage converter (66a), wherein a reference value (18) for the first electrical signal (34) is defined by an absence of X-ray photons (4), wherein for the first electrical signal (34) an upper threshold (20) above the reference value (18) and a lower threshold (22) below the reference value (18) are selected, wherein the first electrical signal (34) is spatially resolved and compared with the upper threshold (20), and if the upper threshold (20) is exceeded, a spatially resolved signal contribution (24a, 24b) of a second electrical signal (36) is generated, wherein the first electrical signal (34) is spatially resolved and compared with the lower threshold (22), and if the lower threshold (22) is undershot, a spatially resolved signal contribution (32) of a correction signal (30) is generated, and wherein the X-ray detector (2) operates in a quantum-counting manner, such that when the upper threshold (20) is exceeded by a signal contribution (14a, 14b) in the first electrical signal (34) in the second electrical signal (36) a natural number, preferably 1, is generated as the value of the generated signal contribution (24a, 24b) and when the lower threshold (22) is undershot by the corresponding signal contribution of the first electrical signal (34) as the signal contribution (32) of the electrical correction signal (30) a constant value, preferably 1, for all possible signal contributions of the first electrical signal (34), wherein the second electrical signal (36) is corrected by the correction signal (30), and an image data set (52) is generated on the basis of the corrected second electrical signal (38), where - the second electrical signal (36) is corrected locally in a pixel electronics (12a, 12b) of the X-ray detector (2) by a simple subtraction of the corresponding spatially resolved signal contributions (24a, 24b) by the correction signal (30), and a third electrical signal (38) is generated from this, or - the second electrical signal (36) and the correction signal (30) are each passed from a pixel electronics (12a, 12b) of the X-ray detector (2) to an application-specific integrated circuit (40, 40a, 40b), and the second electrical signal (36) is corrected locally in the application-specific integrated circuit (40, 40a, 40b) by a simple subtraction of the corresponding spatially resolved signal contributions (24a, 24b) by the correction signal (30), and a third electrical signal (38) is generated from this, and wherein the image data set (52) is generated using the third electrical signal (38).

Description

[0001] The invention relates to a method for correcting a spatially resolved photon measurement of an X-ray detector, wherein a signal contribution in a first electrical signal is generated in the X-ray detector from an incident X-ray photon, wherein an upper threshold is selected for the first electrical signal, wherein the first electrical signal is compared with the upper threshold in a spatially resolved manner, and if the upper threshold is exceeded, a signal contribution of a spatially resolved second electrical signal is generated, and wherein an image data set is generated on the basis of the second electrical signal.

[0002] In a computed tomography (CT) scanner, X-ray images of the patient's body are generated from various polar angles relative to its axis. These individual X-ray images, representing the absorption of X-rays by the body tissue for each angle, are then used to reconstruct a three-dimensional volume model of the body tissue. The spatial resolution of this volume model depends on the one hand on the spatial resolution of the individual X-ray images, i.e., the resolving power of the X-ray detector, and on the other hand on the absorption contrast, i.e., how detailed the varying absorption of the X-rays, which are directed onto the body tissue from the respective angle, can be depicted in a single X-ray image.In order to keep the radiation dose for the patient being examined at a medically acceptable level at all times, an X-ray detection system is advantageous, which is able to sufficiently contrast the different absorption through different tissue layers even with a comparatively low radiation dose in a single X-ray image.

[0003] Quantum-counting X-ray detectors have proven advantageous for achieving both the highest possible spatial resolution and the highest possible contrast in depicting different absorption coefficients, while simultaneously using comparatively moderate X-ray radiation. In a quantum-counting X-ray detector, an incident X-ray photon first generates a free electron in a semiconductor, for example, cadmium telluride, by ionizing one of the lattice atoms. Due to its remaining kinetic energy, this free electron in turn ionizes further lattice atoms, creating an electron-hole pair cloud in the semiconductor. The electrons and holes are then collected by pixelated anodes and cathodes, respectively, to which a bias voltage is applied.When an electron cloud hits an anode pixel, a current pulse is generated first, and from this, via preprocessing, usually a voltage pulse is generated, from which the corresponding X-ray photon can be deduced, which was responsible for the generation of the electron cloud in the area of ​​the anode pixel.

[0004] Due to the effects of image charges, an X-ray photon can, particularly at the boundary between two detector pixels, induce a current or voltage pulse in the adjacent pixel. This means that the X-ray photon, through its charge cloud at the point of impact, generates a corresponding voltage pulse in the detector pixel, and, as a result of the induced charges in the neighboring pixel, another, usually weaker, voltage pulse. This can lead to a deterioration of image contrast due to the counting events, which are caused only by induction from a neighboring area and not by the X-ray photon itself. This can be particularly problematic at high spatial resolution, which is desirable for computed tomography, because the edge area, where such induced charges can occur, is enlarged relative to the total area of ​​a pixel.

[0005] Quantum counting detectors allow a specific minimum energy to be set, below which X-ray photons are not detected at all. To filter out the induction-based counting events—that is, voltage pulses that were actually generated by an X-ray photon at a neighboring pixel—the minimum energy can be raised so high that the induction-based voltage pulses, which are weaker than the voltage pulses of the "correct" counting events, are no longer registered. However, this has the disadvantage that otherwise "correct" voltage pulses with a correspondingly low energy of the generating X-ray photon are also no longer detected, which again degrades the image contrast.

[0006] US patent 2013 / 0168562A1 describes a device for detecting ionizing radiation. In this device, a semiconductor material converts the charges generated by ionization into a signal pulse, among other things, by means of a charge amplifier. By comparing the maximum of the signal pulse with a threshold value, a signal contribution can be distinguished from background noise. A subsequent comparison of the minimum of the signal pulse with another threshold value can also determine whether the signal contribution is an event caused by electromagnetic interference.

[0007] DE 10 2014 204 042 A1 discloses a method for controlling an X-ray detector, which comprises an X-ray-sensitive sensor layer and an arrangement of pixel electrodes connected to the back of the sensor layer, wherein each of the pixel electrodes can be followed by evaluation electronics integrated in an ASIC.

[0008] The invention is based on the objective of providing a method for correcting counting events in spatially resolved photon measurements that arise in a detector pixel not from an incident X-ray photon, but only from an induction from a neighboring pixel. The invention further aims to provide an X-ray detector with which the correction of such counting events is possible. The first-mentioned problem is solved according to the invention by a method for correcting a spatially resolved photon measurement of an X-ray detector, wherein a signal contribution in a first electrical signal is generated in the X-ray detector from an incident X-ray photon by means of a current-voltage converter, wherein a reference value for the first electrical signal is defined by an absence of X-ray photons, wherein an upper threshold value above the reference value and a lower threshold value below the reference value are selected for the first electrical signal, wherein the first electrical signal is compared with the upper threshold value in a spatially resolved manner, and if the upper threshold value is exceeded, a signal contribution of a second electrical signal is generated in a spatially resolved manner, wherein the first electrical signal is compared with a lower threshold value in a spatially resolved manner.and, when the lower threshold is undershot, a spatially resolved signal contribution of an electrical correction signal is generated, wherein the second electrical signal is corrected by the correction signal, and an image data set is generated based on the corrected second electrical signal. Advantageous and partly inventive embodiments are the subject of the dependent claims and the following description.

[0009] A first electrical signal includes, in particular, a voltage signal and a current signal. A signal contribution in the first electrical signal is understood here to mean, in particular, a deviation of the signal value from the reference value. A definition of the reference value by the absence of X-ray photons means, in particular, that the reference value is given by the value of the first electrical signal that the first electrical signal assumes in the total absence of X-ray photons, as can be achieved, for example, by a corresponding, absolutely photon-impermeable block in the relevant spectral range of the X-ray detector.Spatially resolved generation of a signal transmission in the first electrical signal by the incident X-ray photon means in particular that, based on the signal contribution and the corresponding information contained therein, the point of impact on the X-ray detector can be determined at least approximately, whereby a multitude of positions on the X-ray detector are available as possible points of impact.

[0010] In particular, the upper and / or lower threshold values ​​for the first electrical signal can each be selected as spatially dependent, meaning that different values ​​for the upper and lower threshold values ​​can be chosen for signal contributions that differ within the spatial resolution of their generation. However, the upper and lower threshold values ​​can also be chosen to be the same for all possible signal contributions within the first electrical signal, within the spatial resolution. The signal contribution generated in the X-ray detector from the incident X-ray photon, with spatial resolution, contains, in particular, information about the energy of the generating X-ray photon and preferably allows a proportionality to be established with respect to the energy, for example, based on the pulse height of a signal pulse as a signal contribution, or based on a time integral over such a pulse height.

[0011] The spatially resolved generation of a signal contribution in the second electrical signal when the upper threshold is exceeded by a corresponding signal contribution in the first electrical signal can be achieved, in particular, by using only those values ​​of the first signal that exceed the upper threshold. According to the invention, when the upper threshold is exceeded by a signal contribution in the first signal, a natural number, preferably 1, is generated in the second signal as the value of the generated signal contribution. Conversely, when the lower threshold is not reached by the corresponding signal contribution of the first electrical signal, a constant value, preferably 1, is generated as the signal contribution of the electrical correction signal, applicable to all possible signal contributions of the first electrical signal. In this case, the X-ray detector operates as a quantum-counting detector.

[0012] Signal contributions generated in a spatially resolved X-ray detector by an X-ray photon incident on a specific detector pixel typically result in a unipolar, and with appropriate normalization, positive signal pulse within a corresponding current signal, due to the direct charge flow within that pixel. If the current signal is converted into a voltage signal for better processing, it is possible to generate a nonpolar signal pulse within the voltage signal from the unipolar signal pulse in the current signal. To generate a corresponding counting event from such a signal pulse, while maintaining the highest possible signal contrast against potential noise, the upper threshold is chosen to avoid counting events that do not correspond to an X-ray photon incident on the detector pixel.This applies in particular to signal pulses that are not generated by a photon striking the detector pixel itself, but by a photon striking the neighboring detector pixel, thereby causing an induction of charge carriers.

[0013] The signal pulses of such induction-based signal contributions, i.e., those generated by X-ray photons striking the adjacent detector pixel, are initially bipolar in the current signal due to the alternating charge flow of the induction for charge equalization. This means that, within the framework of the corresponding normalization, they first exhibit a positive pulse waveform followed by a decay phase down to a minimum below the value corresponding to the absence of X-ray photons, before subsequently returning to this value. Accordingly, a voltage pulse with a bipolar waveform can also be generated from such a bipolar current pulse. This can be achieved, for example, using a current-to-voltage converter, such as a transimpedance amplifier.

[0014] Depending on whether the first electrical signal is a current or voltage signal, the signal pulse exhibits a minimum below the reference value before returning to the reference value. According to the procedure, such signal pulses are removed from the count—since they do not objectively correspond to any X-ray photon striking the detector pixel—by having the undershooter, with its pulse minimum, contribute to the correction signal. This allows for the simple correction of the corresponding counting events in the second electrical signal that are based on induction rather than an actual X-ray photon.

[0015] According to the invention, the second electrical signal is corrected by a simple subtraction of the corresponding spatially resolved signal contributions by the correction signal.

[0016] Advantageously, the signal contribution of the first electrical signal is generated by quantum counting. The described method is particularly advantageous for a quantum-counting X-ray detector, since simple counting events can be offset against each other.

[0017] Preferably, the ratio of the lower threshold value to the upper threshold value, both relative to the reference value, is chosen to be between 0.3 and 3, and particularly preferably between 0.5 and 2. This means that the difference between the upper threshold and the reference value, on the one hand, and the difference between the lower threshold and the reference value, on the other, have a ratio to each other that falls within the specified interval. This ensures, firstly, that the corresponding sub-vibrations are correctly registered and lead to corresponding signal contributions in the correction signal, and secondly, that the upper threshold is kept sufficiently low so that even low-energy X-ray photons can be detected.This leads to an advantageous signal contrast, since, due to the higher absorption rate of low-energy X-ray photons in body tissue, the information on the transmission of these X-ray photons is particularly valuable.

[0018] Advantageously, the lower threshold is chosen in an interval of 5 keV to 15 keV, particularly preferably between 5 keV and 10 keV below the reference value. This means that the energy of an incident X-ray photon is taken into account by the respective normalization of the first electrical signal in such a way that the energy of the generating X-ray photon can be determined from the first electrical signal. The lower threshold is then chosen such that the energy of an X-ray photon corresponding to the threshold lies within the aforementioned interval. This allows signal contributions in the correction signal due to electronic noise in the first electrical signal to be suppressed, and also ensures that as many sub-oscillators of bipolar signal pulses as possible are detected, leading to the correction of a particularly large number of influence-based, and not "X-ray photon-generated," signal contributions in the individual detector pixels.

[0019] According to the invention, the second electrical signal is corrected locally in the pixel electronics of the X-ray detector by the correction signal, and a third electrical signal is generated from this correction. The image data set is then generated based on this third electrical signal. This means, in particular, that the second electrical signal, in its entirety with the corresponding spatial information relating to the corresponding generation points of the first electrical signal in the X-ray detector, is corrected signal by signal by signal directly locally in the pixel electronics by the correction signal, in its entirety with the corresponding spatial information relating to the points where the lower threshold value is undercut by the first electrical signal. The pixel electronics in this context include, in particular, electrodes for collecting charge carriers and for the corresponding generation of a current signal.

[0020] Preferably, the X-ray detector is configured to correct the corresponding signal contributions in the second electrical signal and the correction signal directly within the individual detector pixels, before further processing, such as that which occurs in an application-specific integrated circuit (ASIC) connected to the pixel electronics. This generates the third electrical signal, which is then further processed in the respective ASICs connected to the individual pixel electronics, for example, by appropriate signal amplification. Local and, in particular, pixel-by-pixel generation of the third electrical signal has the advantage that only the latter needs to be processed further, thus saving power and computational resources.

[0021] The correction can be achieved, for example, by selecting the upper and lower threshold values ​​only as high as necessary to suppress unwanted quantum noise, and ensuring that all signal contributions in the first electrical signal are converted into corresponding signal contributions in the second electrical signal or the correction signal. In a preferred embodiment, positive current pulses of the second electrical signal can then be combined with negative current pulses of the correction signal to generate the third electrical signal. The actual counting events can then be determined from the resulting third electrical signal.

[0022] In an alternative embodiment of the invention, the second electrical signal and the correction signal are each passed from pixel electronics of the X-ray detector to an ASIC, wherein the second electrical signal is corrected locally within the ASIC by the correction signal, and a third electrical signal is generated from this, and wherein the image data set is generated based on the third electrical signal. By correcting the second electrical signal outside the actual pixels of the X-ray detector, it can be ensured that in the event of a problem or defect in the circuit implementing the correction, the sensitive components of the X-ray detector responsible for generating the first electrical signal are not affected by measures to rectify the defect.Because the correction takes place within the ASIC, which is usually directly connected to pixel electronics, the correction is still sufficiently local to require only one signal path for the third electrical signal for further processing.

[0023] Alternatively, the second electrical signal and the correction signal can each be passed from an ASIC to a central signal processor, whereby the second electrical signal is corrected by the correction signal in the central signal processor, and a third electrical signal is generated from this, whereby the image data set is generated based on the third electrical signal.This includes, in particular, that the second electrical signal, in the totality of its signal contributions with the corresponding spatial information relating to the corresponding generation points of the first electrical signal in the X-ray detector, is first transmitted signal contribution by signal contribution from the individual ASICs, each of which is assigned to one or a small group of detector pixels, to the central signal processing unit, and the correction by the correction signal, in the totality of its signal contributions with the corresponding spatial information relating to the deviations of the lower threshold value by the first electrical signal, takes place in the central signal processing unit.

[0024] Correction in a central signal processing unit outside the ASIC has the advantage that the individual ASICs only need to be designed for a minimum of the functions necessary for signal conditioning and amplification, and do not need to be additionally designed and supplied with power for correction.

[0025] Alternatively, the second electrical signal and the correction signal can each be transmitted from a central signal processor on a rotating part of the CT scanner to a stationary part of the CT scanner. The second electrical signal is then corrected by the correction signal on the stationary part of the CT scanner, and a third electrical signal is generated from this correction. The image data set is then generated based on this third electrical signal. Correcting the second electrical signal on the stationary part of the CT scanner has the advantage that no additional processing power needs to be transferred to the rotating part for the logical operations that implement the correction.

[0026] Alternatively, the second electrical signal and the correction signal can each be transmitted to an image processing unit, wherein a preliminary image data set is generated from the second electrical signal, a correction image data set is generated from the correction signal, and the image data set is generated from the preliminary image data set and the correction image data set. This includes, in particular, the X-ray detector being arranged on the rotating part of a CT scanner and the image processing unit being arranged on the stationary part of a CT scanner.Generating the image data set using a preliminary image data set generated from the second electrical signal and a correction image data set generated from the correction signal has the advantage that the image processing unit's infrastructure, which is already available for generating image data, can be used to minimize the influence of the counting events represented in the correction signal's signal contributions on the final image data set.

[0027] The specific embodiment of the correction of the second electrical signal and / or the generation of the image data set can be chosen depending on further requirements, which may arise in particular from the X-ray detector to be used and / or from the higher-level application such as a CT scanner.

[0028] The second problem is solved according to the invention by an X-ray detector for the spatially resolved detection of X-ray photons, which is configured to generate a spatially resolved signal contribution in a first electrical signal from an incident X-ray photon, for which a reference value is defined by the absence of X-ray photons, and to resolve both positive and negative signal contributions of the first electrical signal with respect to the reference value and to make them available for further processing. The advantages stated for the method and for its further developments can be transferred analogously to the X-ray detector.

[0029] According to the invention, the X-ray detector comprises a converting detector material for generating charge carriers from an incident X-ray photon, a spatially resolved electrode arrangement, and a plurality of current-to-voltage converters. The X-ray detector is further configured to generate a spatially resolved current pulse from the charge carriers generated by an incident X-ray photon using the electrode arrangement, and to generate a voltage pulse as a signal contribution to the first electrical signal using one of the current-to-voltage converters based on the current pulse.

[0030] Preferably, the converting detector material is a semiconductor or a suitably designed ceramic. In particular, the electrode arrangement is in direct contact with the converting detector material and is configured to collect the charge carriers generated in the detector material by an incident X-ray photon via an applied voltage, thereby generating the current pulse. Depending on the applied voltage and the charge carriers to be collected, the arrangement is either a cathode or anode. The choice of charge carriers to be collected and the corresponding applied voltage is preferably determined based on the charge carrier mobility in the converting detector material.

[0031] For spatial resolution, the electrode arrangement is divided into individual electrode pixels, so that each current-to-voltage converter is connected to a small number, preferably exactly one, electrode pixel. The respective current-to-voltage converter then receives a current pulse generated in the corresponding electrode pixel from the collected charge carriers and converts this into a voltage pulse, which is further processed as a signal contribution in the first electrical signal in the manner described above.

[0032] According to the invention, the current-to-voltage converters are each configured to generate a unipolar voltage pulse from a unipolar current pulse generated by an incident X-ray photon, and to generate a bipolar voltage pulse from a bipolar current pulse. Preferably, the current-to-voltage converter is operated such that a monotonic relationship exists between current pulses generated by X-ray photons of different energies and the resulting voltage pulses. That is, the energy of the incident X-ray photon can be deduced from the maximum pulse amplitude of the resulting voltage pulse or from a time integral over the voltage pulse. This is particularly advantageous for further processing, especially with regard to correcting induction-based counting events.An additional advantage of such voltage pulses is that, due to the monotony, the corresponding voltage levels can be implemented using comparators.

[0033] In the concrete implementation of the current-to-voltage converter, it is important to ensure that the conversion reacts quickly enough to changes in the current pulse so that the different contributions of bipolar current pulses, which are generated from induced charges, are not "integrated" into a unipolar voltage pulse during the conversion process. Furthermore, self-excited oscillations in the voltage pulse must be avoided when converting a unipolar current pulse. These oscillations could arise, for example, from an insufficient phase margin in the current-to-voltage converter. To achieve this, the current-to-voltage converter can be operated with a largely linear characteristic, and a suitable filter, adapted to the aforementioned requirements, can be connected downstream. This filter might be a passive RC circuit.However, the current-to-voltage converter can also be adapted to these requirements with regard to its own transfer function, e.g. by choosing an appropriate feedback impedance in the converter or similar.

[0034] According to the invention, the current-to-voltage converters are each integrated into an application-specific integrated circuit (ASIC), which each includes an amplifier stage. Preferably, the individual current-to-voltage converters are themselves formed by an amplifier stage, e.g., in the form of a transimpedance amplifier. Designing the current-to-voltage converter as an amplifier stage has the advantage that, on the one hand, the generated current pulses have a comparatively low signal amplitude for conventional signal processing, while the conversion into a voltage pulse can be accompanied by an amplification of the signal amplitude. On the other hand, the pulse shape of the voltage pulses generated from the current pulses can be controlled particularly easily via a transfer function of the amplifier stage or via a filter connected downstream of the amplifier stage.Integrating the current-to-voltage converter into an ASIC has the additional advantage that the circuit can be specifically optimized for generating voltage pulses with the aforementioned properties.

[0035] The invention further describes a medical imaging device comprising an X-ray source for generating X-ray photons and a previously described X-ray detector, which is configured to carry out the previously described method. The advantages stated for the method and its further developments, as well as for the X-ray detector and its further developments, can be applied analogously to the medical imaging device.

[0036] Preferably, the medical imaging device is a computed tomography scanner. Given the high demands on image quality, and especially on the contrast of the individual images used in a computed tomography scanner to generate the volume model of the tissues under investigation, the use of the aforementioned method in conjunction with the aforementioned X-ray detector is particularly advantageous.

[0037] Advantageously, the medical imaging device is designed as a C-arm device. The aforementioned procedure and the aforementioned X-ray detector also prove to be particularly advantageous for such devices and the image quality requirements they impose.

[0038] An embodiment of the invention is explained in more detail below with reference to a drawing. The drawing schematically depicts: Fig. 1. An X-ray photon striking the detector pixels of an X-ray detector in an oblique view. Fig. 2 voltage pulses generated in the detector pixels by an X-ray photon in a diagram according to Fig. 1, Fig. 3 the generation and corresponding correction of counting events in the voltage pulses according to Fig. 2, Fig. 4 a CT scan with an X-ray detector, in which the correction according Fig. 3 in pixel electronics, Fig. 5 a CT scan with an X-ray detector, with the correction according to Fig. 3 in one of the ASICs downstream of the pixel electronics, Fig. 6 a CT scan with an X-ray detector, with the correction according to Fig. 3 in a central signal processing system downstream of the ASICs, Fig. 7 a CT scan with an X-ray detector, with the correction according to Fig. 3. after data transmission on the stationary part, Fig. 8 a CT scan with an X-ray detector, with the correction according to Fig. 3 via a correction image generated from the correction signal, and Fig. 9 in an oblique view an X-ray detector for generating the voltage pulses according to Fig. 2 and Fig. 3.

[0039] Corresponding parts and sizes are marked with the same reference symbols in all figures.

[0040] In Fig. Figure 1 schematically depicts detector pixels 1a, 1b, and 1c of a quantum-counting X-ray detector 2 in an oblique view. An X-ray photon 4 incident in the region of pixel 1a first generates a charge cloud 8 in a semiconductor crystal 6 through ionization. Due to a corresponding bias voltage applied in the region of detector pixel 1a, the charge cloud 8 diffuses to the pixel electronics 12a, which includes a corresponding anode 10a. In the pixel electronics 12a, the charge cloud 8, caused by the X-ray photon 4, generates a current pulse, which is processed into a voltage pulse in an ASIC (not shown). However, due to induction, a corresponding current pulse is also generated in the pixel electronics 12b, which is associated with detector pixel 1b, and this is processed into a voltage pulse. These voltage pulses are each represented in Fig. 2 shown.

[0041] In Fig. Figure 2 in a diagram are the voltage pulses 14a, 14b, which are determined based on the values ​​in the detector pixels 1a, 1b. Fig. The current pulses generated by the X-ray photon 4 were plotted against time t. The voltage U for voltage pulse 14a is proportional to the energy of the incident X-ray photon. In this case, the respective current pulse was processed using a suitable current-to-voltage converter such that the unipolar positive current pulse generated by the X-ray photon 4 in detector pixel 1a generates the unipolar voltage pulse 14a, while the bipolar current pulse generated by the X-ray photon 4 in detector pixel 1b generates the bipolar voltage pulse 14b.

[0042] It is clearly evident that the peak of voltage pulse 14b lies less than an order of magnitude below the peak of voltage pulse 14a. If voltage pulse 14a originates from a particularly high-energy X-ray photon, and if the point of impact in pixel 1a is particularly close to the edge of the neighboring pixel 1b, then the peak 16b of voltage pulse 14b may well be of a magnitude comparable to that of independent counting events of lower-energy X-ray photons. To prevent voltage pulse 14b from resulting in a corresponding counting event that contributes to the X-ray image, a correction is applied based on the Fig. 3 is described.

[0043] In Fig. 3 are the voltage pulse 14a and the voltage pulse 14b according to Fig. For clarity, the values ​​are plotted against time t in separate diagrams. The time range shown corresponds to the minimum time resolution of the X-ray detector. The voltage value of 0 V is chosen such that this value is present in a corresponding signal when X-ray photons are absent. The value of 0 V is henceforth considered the reference value 18. However, this is only one possible choice for a reference value. The reference value can also be chosen, for example, by a corresponding linear scaling of a voltage range between the supply voltage and the ground of a transimpedance amplifier generating the first electrical signal, where ground is assigned the value of 0 V.

[0044] An upper threshold 20 and a lower threshold 22 are now chosen consistently for both voltage pulses 14a and 14b. Preferably, the upper threshold 20 and the lower threshold 22 are chosen to be the same for all detector pixels of a detector. This simplifies the correction, as additional assumptions can be avoided due to the consistency across the detector. As a consequence of the voltage pulse 14a exceeding the upper threshold 20, a corresponding counting event 24a is generated for pixel 1a in an ASIC connected to the pixel electronics 12a. As shown, the duration of the counting event 24a does not necessarily have to correspond to the duration of the voltage pulse 14a exceeding the upper threshold 20, but can be generated as a normalized on / off signal.

[0045] In detector pixel 1b, a counting event 24b is also generated as a result of the peak 16b of the voltage pulse 14b exceeding the upper threshold 20. However, since the voltage pulse 14b also exhibits an undershoot 26 due to the induced charge, this can be adjusted by appropriately choosing the lower threshold 22 between the reference value 18 and the negative peak 28 of the undershoot 26, such that a voltage pulse 14b with sufficient signal amplitude also leads to a corresponding counting event 32 in the correction signal 30. The counting events 24b in the second electrical signal and 32 in the correction signal 30 are so close together in time that they are perceived as simultaneous within the temporal resolution of the count rate – which is represented by the depicted range.

[0046] This now forms the basis for a correction: For the final generation of the images, only those counting events 24a are used for which, within the temporal resolution, no corresponding counting event 32 is present in the correction signal 30 at the same pixel. This ensures that it is a unipolar voltage pulse of type 14a, and not a bipolar voltage pulse of type 14b. Due to the pulse shaping of the first electrical signal as a voltage signal from the current signals of the detector pixels 1a-1c, which always converts unipolar current pulses into unipolar voltage pulses, and thus bipolar voltage pulses can only originate from bipolar current pulses, it can be concluded that the corresponding counting event 24a was indeed caused by an X-ray photon, and not by the induction of charges that were originally generated by an X-ray photon at the neighboring pixel.

[0047] There are several possibilities for the actual correction of the image data set generated from the second signal using the correction signal 30. These are illustrated in the following figures.

[0048] In Fig. Figure 4 schematically depicts a quantum-counting X-ray detector 2 in a block diagram, in which an impact of the X-ray photon 4 initially leads to a signal contribution in a first electrical signal 34, which is converted in the pixel electronics 12a into a signal contribution in a second electrical signal 36. Still in the pixel electronics 12a, the signal contributions for the correction signal 30 are derived from the first electrical signal 34 according to Fig. 3 is generated, and the second electrical signal 36 is locally corrected in the pixel electronics 12a by the correction signal 30. This can be done, for example, by removing the negative signal contributions of the current pulse from the positive signal contributions of a bipolar current pulse in the first electrical signal.

[0049] The resulting third electrical signal 38 is amplified in an ASIC 40, and subsequently, the individual signal contributions of the third electrical signal 38 are passed from the various ASICs 40 to a central signal processing unit. From the central signal processing unit 42, which is also referred to as the "module back plane" and which is arranged on the rotating part 44 of the CT 46, the processed data is transmitted to the stationary part 48 of the CT 46, and an image data set 52 is generated there in an image processing unit 50.

[0050] In Fig. 5 shows a possible correction, which is an alternative to the one based on Fig. The process shown in Figure 4 is as follows. Here, the X-ray photon 4 strikes the quantum-counting X-ray detector 2, generating a current pulse 33 in the pixel electronics 12a. This pulse is first converted into the first electrical signal 34 in the ASIC 40, from which the signal contributions of the second electrical signal 36 and the correction signal 30 are generated. The actual correction is performed in the ASIC 40 based on the signal contributions of the second electrical signal 36 and the correction signal 30. The resulting third electrical signal 38 is output to the central signal processing unit 42 and from there transmitted to the stationary part 48 of the CT scanner 46, where the image data set 52 is generated in the image processing unit 50.

[0051] Another alternative for correcting the in Fig. 4 and Fig. The 5 shown are based on Fig. Figure 6 is shown schematically. In the circuit diagram shown there, the second electrical signal 36, whose counting events or signal contributions include the actual photon counts as well as possibly influence-based phenomena, and the correction signal, whose counting events include exclusively insolvency-based phenomena, are each passed from the ASIC 40 to the central processing unit 42, where the actual correction takes place. The generated third electrical signal 38 is then passed from the central signal processing unit 42 on the rotating part 44 of the CT 46 to its stationary part 48, and the image data set 52 is generated from this in the image processing unit 50.

[0052] However, the correction of the counting events can be made to the fixed part 48 of CT 46. This is based on Fig. 7 and Fig. 8 shown. In the embodiment which does not belong to the core of the present invention, but may be inventive in itself, according to Fig. 7. The second electrical signal 36, i.e., the pixel-wise counting events above the upper threshold, and the correction signal 30, i.e., the pixel-wise counting events below the lower threshold, are each transmitted separately to the central signal processing unit 42. The signals then comprise the counting events of all detector pixels. From the central signal processing unit 42, both the second electrical signal and the correction signal, possibly after appropriate post-processing, are still transmitted separately from the rotating part 44 to the stationary part 48 of the CT 46.

[0053] In the exemplary embodiment according to Fig. 7. The correction of the second electrical signal 36 by the correction signal 30, and the corresponding generation of the third electrical signal 38, takes place before the image processing unit. The image processing unit 50 then generates the image data set 52 from the corrected, i.e., the third electrical signal 38.

[0054] In the embodiment which does not belong to the core of the invention, but may be inventive in itself, according to Fig. 8. Both the second electrical signal 36 and the correction signal 30 are transmitted to the image processing unit 50. A preliminary image data set 54 is generated from the second electrical signal, and a correction image data set 56 is generated from the correction signal 30. The final image data set 52 is then generated by correcting the preliminary image data set 54 by the correction image data set 56.

[0055] In Fig.Figure 9 schematically depicts an X-ray detector 2. The X-ray detector 2 has a semiconductor crystal 6 as its converting detector material. The electrons generated in the semiconductor crystal 6 by an incident X-ray photon are collected with spatial resolution by an electrode arrangement comprising the anode pixels 10a, 10b. For this purpose, a bias voltage 57a, 57b is applied to each of the anode pixels 10a, 10b.

[0056] The anode pixels 10a and 10b are each connected to an ASIC 40a and 40b, respectively. In the ASIC 40a, a current pulse 33a generated by the anode pixel 10a is fed to an amplifier stage 58a, which in this case is designed as a transimpedance amplifier. The amplifier stage 58a has a supply voltage 60a and a feedback element 62a with a resistor 64a, through which the resulting first electrical signal 34 is fed back into the current pulses 33a generated by the anode pixel 10a. The amplifier stage 58a thus operates as a current-to-voltage converter 66a, which spatially resolves the current pulse 33a and generates a corresponding voltage pulse 14a in the first electrical signal 34. The further processing of the first electrical signal 34 then takes place in the manner described above.

[0057] Although the invention has been illustrated and described in detail by means of the preferred embodiment, the invention is not limited by this embodiment. Other variations can be derived from it by a person skilled in the art without departing from the scope of protection of the invention.

Claims

[1] Method for correcting a spatially resolved photon measurement of an X-ray detector (2), wherein in the X-ray detector (2) a signal contribution (14a, 14b) in a first electrical signal (34) is generated in a spatially resolved manner from an incident X-ray photon (4) by means of a current-voltage converter (66a), wherein a reference value (18) for the first electrical signal (34) is defined by an absence of X-ray photons (4), wherein for the first electrical signal (34) an upper threshold value (20) above the reference value (18) and a lower threshold value (22) below the reference value (18) are selected, wherein the first electrical signal (34) is compared with the upper threshold value (20) in a spatially resolved manner, and when the upper threshold value (20) is exceeded, a signal contribution (24a, 24b) of a second electrical signal (36) is generated in a spatially resolved manner, wherein the first electrical signal (34) is compared with the lower threshold value (22) in a spatially resolved manner, and when the lower threshold value (22) is undershot, a signal contribution (32) of a correction signal (30) is generated in a spatially resolved manner, and wherein the X-ray detector (2) operates in a quantum counting manner, so that when the upper threshold value (20) is exceeded by a signal contribution (14a, 14b) in the first electrical signal (34) in the second electrical signal (36), a natural number, preferably 1, is generated as the value of the generated signal contribution (24a, 24b), and when the lower threshold value (22) is undershot by the corresponding signal contribution of the first electrical signal (34), a constant value, preferably 1, is generated for all possible signal contributions of the first electrical signal (34) as the signal contribution (32) of the electrical correction signal (30), wherein the second electrical signal (36) is corrected by the correction signal (30), and an image data set (52) is generated based on the corrected second electrical signal (38), where - the second electrical signal (36) is corrected locally in a pixel electronics unit (12a, 12b) of the X-ray detector (2) by a simple subtraction of the corresponding spatially resolved signal contributions (24a, 24b) by the correction signal (30), and a third electrical signal (38) is generated therefrom, or - the second electrical signal (36) and the correction signal (30) are each forwarded by a pixel electronics (12a, 12b) of the X-ray detector (2) to an application-specific integrated circuit (40, 40a, 40b), and the second electrical signal (36) is corrected locally in the application-specific integrated circuit (40, 40a, 40b) by a simple subtraction of the corresponding spatially resolved signal contributions (24a, 24b) by the correction signal (30), and a third electrical signal (38) is generated therefrom, and wherein the image data set (52) is generated based on the third electrical signal (38). [2] Method according to claim 1, wherein the signal contribution (14a, 14b) of the first electrical signal (34) is generated by quantum counting. [3] Method according to claim 1 or claim 2, wherein the ratio of the amount of the lower threshold value (22) and the amount of the upper threshold value (20), in each case relative to the reference value (18), is selected between 0.3 and 3. [4] Method according to one of the preceding claims, wherein the lower threshold value (22) is selected in an interval of 5 keV to 15 keV below the reference value (18). [5] X-ray detector (2) for the spatially resolved detection of X-ray photons (4), comprising - a converting detector material (6) for generating charge carriers from an incident X-ray photon (4), - a spatially resolving electrode arrangement (10a, 10b) and - a plurality of current-voltage converters (66a), each integrated into an application-specific integrated circuit (40, 40a, 40b), each comprising an amplifier stage (58a), wherein the X-ray detector (2) is configured to - to generate a spatially resolved signal contribution (14a, 14b) in a first electrical signal (34) from an incident X-ray photon (4), for which a reference value (18) is defined by an absence of X-ray photons (4), - to generate a spatially resolved current pulse (33, 33a) in the first electrical signal (34) from the charge carriers generated by an incident X-ray photon (4) by means of the electrode arrangement (10a), - to generate a voltage pulse (14a, 14b) as a signal contribution (14a, 14b) in the first electrical signal (34) based on the current pulse (33, 33a) by means of one of the current-voltage converters (66a), - to resolve both positive signal contributions (14a, 14b, 16b) with respect to the reference value (18) and negative signal contributions (26, 28) of the first electrical signal (34) with respect to the reference value and to make them available for further processing, wherein the current-voltage converters (66a) are each configured to generate a unipolar voltage pulse (14a) based on a unipolar current pulse (33, 33a) generated by an incident X-ray photon (4), and to generate a bipolar voltage pulse (14b) based on a bipolar current pulse (33, 33a), and wherein for the first electrical signal (34) an upper threshold value (20) above the reference value (18) and a lower threshold value (22) below the reference value (18) are selected, and the X-ray detector (2) is also arranged - to compare the first electrical signal (34) with the upper threshold value (20) in a spatially resolved manner, and to generate a signal contribution (24a, 24b) of a second electrical signal (36) in a spatially resolved manner when the upper threshold value (20) is exceeded, - to compare the first electrical signal (34) with the lower threshold value (22) in a spatially resolved manner, and to generate a signal contribution (32) of a correction signal (30) in a spatially resolved manner when the lower threshold value (22) is undershot, - when the upper threshold value (20) is exceeded by a signal contribution in the first electrical signal (34), to generate a natural number, preferably 1, in the second electrical signal (36) as the value of the generated signal contribution, - if the corresponding signal contribution of the first electrical signal (34) falls below the lower threshold value (22), to generate as the signal contribution (32) of the electrical correction signal (30) a constant value, preferably 1, for all possible signal contributions of the first electrical signal (34), and - correcting the second electrical signal (36) by a simple subtraction of the corresponding spatially resolved signal contributions (24a, 24b) in order to create the correction signal (30). [6] Medical imaging device comprising an X-ray source for generating X-ray photons (4) and an X-ray detector (2) according to claim 5, which is arranged to carry out the method according to one of claims 1 to 4. [7] Medical imaging device according to claim 6, which is designed as a computer tomograph (46). [8] Medical imaging device according to claim 6, which is designed as a C-arm device.

Images