Compensation for Polarization Effects in Photon Counting Detectors

Abstract

The present invention relates to a system and method for correcting the number of counts in an energy bin of X-ray photons detected by a photon counting detector for a spectral computed tomography system, wherein the illumination history of the photon counting detector is taken into account to determine a gain and / or offset of the photon counting detector, and the number of counts in an energy bin of the detected photons is corrected according to a correction value corresponding to the determined gain and / or offset.

Description

Technical Field

[0001] The present invention relates to a system for correcting the number of counts in an energy bin of X-ray photons detected by a photon-counting detector for a spectral computed tomography system, and a method for correcting the number of counts in an energy bin of X-ray photons detected by a photon-counting detector for a spectral computed tomography system.

Background Art

[0002] A photon-counting detector counts photons that impinge on the detector and determines their energies. The photon-counting detector primarily provides a signal having a pulse height that can be proportional to the energy of the incoming photons. Such a normal reconstruction of photon energy can only be applied when the count rate of the incoming photons is relatively low. For example, due to the high X-ray intensity used in medical imaging systems such as computed tomography, the proportion of photons impinging on the detector becomes very high. These high proportions of detected photons saturate the detector. In the case of a semiconductor detector, polarization of the photon-counting detector occurs, as a result of which the signal of the detector becomes low and the pulse height of the signal decreases. Thereby, the reconstructed energy of the detected photons decreases and may be classified into incorrect energy bins of the detection system. This effect can cause ring or band artifacts in computed tomography images, depending on the predominance of this effect at the pixel level or the module / crystal level.

[0003] For these reasons, it would be advantageous to have a system and method for correcting the number of counts in an energy bin of X-ray photons detected by a photon-counting detector for a spectral computed tomography system that is not affected by the above-mentioned drawbacks and can reduce polarization artifacts in photon-counting computed tomography image reconstruction.

[0004] WO2017 / 046002A1 discloses photon count correction in a photon counting X-ray radiation detection system.

[0005] EP3567405A1 discloses a radiation detector adapted to detect leakage current.

[0006] US2012 / 243660A1 discloses a method for correcting detector data of an X-ray detector and an X-ray recording system.

SUMMARY OF THE INVENTION

PROBLEMS TO BE SOLVED BY THE INVENTION

[0007] An object of the present invention is to provide a system and method for correcting the number of counts in the energy bins corresponding to a non-polarized detector for the number of counts in the energy bins of X-ray photons detected by a photon counting detector for a spectral computed tomography system.

MEANS FOR SOLVING THE PROBLEMS

[0008] The object of the present invention is solved by the subject matter of the independent claims, and further embodiments are incorporated into the dependent claims.

[0009] The described embodiments also relate to a system for correcting the number of counts in the energy bins of X-ray photons detected by a photon counting detector for a spectral computed tomography system, and a method for correcting the number of counts in the energy bins of X-ray photons detected by a photon counting detector for a spectral computed tomography system. Synergistic effects may arise from different combinations of embodiments, although they may not be described in detail.

[0010] Furthermore, it should be noted that although all embodiments of the present invention related to the method are executed in the order of the described steps, this does not necessarily have to be the only and essential order of the method steps. The methods presented herein can be executed in a different order of the disclosed steps without departing from each respective method embodiment, unless explicitly stated otherwise below.

[0011] According to a first aspect of the present invention, there is provided a system for correcting the number of counts in the energy bin of X-ray photons detected by a photon-counting detector for a spectral computed tomography system. The system includes a photon detection unit configured to include a photon-counting detector, detect photons, and provide the pulse height of each of the detected photons, and a storage device configured to store the number of counts in the energy bin according to the pulse height of each of the detected photons. The system further includes a determination unit configured to determine a correction value for the gain and / or an offset correction value of the photon-counting detector based on the illumination history of the photon-counting detector, and a correction unit configured to correct the number of counts in the energy bin of the detected photons according to the correction value for the gain and / or the offset correction value of the photon-counting detector. The correction unit is configured to correct the number of counts in the energy bin of the detected photons in the forward model of the spectral computed tomography system by adapting the response function of the photon detection unit according to the correction value. The system is configured to correct the deviation of the number of counts in the energy bin caused by the space charge effect, polarization effect, and / or charge trapping in the photon-counting detector.

[0012] The photon detection unit comprises a photon counting detector. The photon detection unit can further comprise readout electronics for the photon counting detector, such as an amplifier, a shaper, a comparator, a counter, or an analog-to-digital converter. This detector can be, for example, a semiconductor detector made from silicon, cadmium zinc telluride, gallium arsenide, germanium, or cadmium telluride. Photons can interact with the detector material via the photoelectric effect or the Compton effect. The electrons generated by this interaction can accumulate their energy in the active region of the photon counting detector, thereby generating charge carriers such as electrons and holes. These charge carriers can be accelerated towards the anode and cathode of the detector, respectively, by applying an electric field to the detector. However, the separation of the negative and positive charge carriers generated in the electric field can lead to the accumulation of space charge and the possibility of charge trapping. This can depend on the presence of impurities in the detector material, changing the electric field and thus the response behavior of the photon counting detector with respect to the gain and offset of the photon counting detector. When the X-ray intensity is high and thus the proportion of detected photons is high, the response of the photon counting detector can depend on the illumination history of the impinging radiation. As the X-ray intensity increases, the accumulation of space charge and charge trapping in the semiconductor material can cause changes in the electric field. This causes different responses of the detector to a single detected photon depending on the history of the impinging radiation. The response of the photon counting detector as a result of the energy of the incoming photons can be characterized by the gain and offset. The gain of the photon counting detector can be a measure of the number of charge carriers collected at the detector electrodes in relation to the number of charge carriers generated by the incident photons. Assuming no effect of permanent radiation damage on the photon counting detector, after the detector has not been exposed to X-ray irradiation for a sufficiently long time, the initial conditions of the electric field are re-established inside the bulk of the detector. The time scales on which these effects occur can have multiple physical origins, leading to a challenging model of this effect.The system according to the present invention is based on the correction of the deviation of the number of photons detected in the energy bins, whereby a reproducible response of the detector is obtained. The correction is based on the inventor's insight that the output of the detector at any given instant of its operation is determined by the incident spectrum, flux, and illumination history.

[0013] Accordingly, the system of the present invention comprises a photon detection unit configured to detect photons and provide the pulse height of the detected photons, and a storage device configured to store the count numbers in the energy bins according to the respective pulse heights of the detected photons. The system can comprise several energy bins, preferably four or five energy bins, where the count numbers in all bins are corrected. An illumination history is provided which is a record of the photons previously detected in the photon count detector. The illumination history can include the flux of the photons and the time during which the photon count detector is exposed to this photon flux. The flux of the photons to the photon count detector can depend on the position of a particular detection element in a spectral computed tomography system and can depend on the measured projection. The system further comprises a determination unit configured to determine a correction value for the gain and / or offset of the photon count detector based on the illumination history. The determination unit determines the correction value taking into account the preceding photons of the illumination history illuminating the photon detection unit. This correction value can correspond to a decrease or increase in the gain of the photon detection unit. The correction factor can further correspond to an offset of the pulse height of the photon detection unit. The offset can be an extrapolation of the detector's response with respect to the photon energy at which the pulse height becomes zero. The system further comprises a correction unit configured to correct the count numbers in the energy bins of the detected photons according to the correction value. Accordingly, the count numbers in the energy bins are corrected to the count numbers that would have been measured in an unpolarized detector.

[0014] A photon counting detector can serve as a detector for a computed tomography system that measures photons at energies in the X-ray energy range. Therefore, to derive the attenuation coefficient and material composition of an object being examined, the count numbers in one or more energy bins, preferably four or five bins, are used. The count numbers can be corrected directly in the energy bins. However, the count numbers can also be corrected by providing a forward model for predicting the value of the number of X-ray photons detected in the energy bins that account for the polarization state of the detector. The forward model can include the relationship between the attenuation coefficient, and material composition, and the count numbers in the energy bins, and the response function of the photon detection unit. By correcting the pulse height of the photons in the response function with the correction values of the gain and / or offset of the photon counting detector, the material composition can be derived by this indirect correction of the count numbers in the energy bins.

[0015] For example, the measured count numbers of a photon counting detector can be used to determine the material composition of an object penetrated by X-rays. In one embodiment of the present invention, the measured count numbers are corrected for the polarization effect, and then the corrected count numbers and the forward model of a non-polarized detector are used to estimate the material composition of the object. In another embodiment of the present invention, the correction of the count numbers is performed by directly estimating the material composition of the object using the forward model of a polarized detector.

[0016] In an embodiment of the present invention, the illumination history includes the flux of photons previously detected by a photon counting detector at time intervals.

[0017] In this embodiment of the present invention, the illumination history can include the flux and exposure time to which the detector was exposed to radiation. The illumination history can further include the pulse height and time stamp of the detected photons. The pulse height can be a measure of the number of charge carriers reaching the electrodes of the photon-counting detector. The time of the detected photons can include the specific time stamp of the photons reaching the detector. In this case, the photon detection unit can operate in list mode, thereby providing the energy and time of all individual photons. However, the time of the detected photons can include the time interval during which the detector is measuring a specific projection of the computed tomography system. As a result, multiple photons can be stored in the illumination history at the same detection time. In this embodiment of the present invention, the illumination history can include signals of multiple photons previously detected by the photon-counting unit.

[0018] By correcting the count number in the energy bin, the count number is corrected to a value that would have been measured in a detector that was not previously exposed to irradiation. The correction can take into account the changed gain of the photon detection unit. The decrease in gain can be due to space charge effects, polarization effects, or charge trapping in the photon detection unit, thereby resulting in a lower pulse height of the signal than in a photon-counting detector that was not previously exposed to illumination. By correcting the gain with a correction value, a reproducible and reliable energy can be assigned to the detected photons, thereby improving the image quality of the computed tomography system and reducing image artifacts.

[0019] In an embodiment of the present invention, the determination unit comprises an artificial intelligence module configured to determine a correction value for the gain and / or a correction value for the offset of the photon-counting detector based on the illumination history.

[0020] The count number can be corrected using an artificial intelligence module. The artificial intelligence module can be equipped with a recurrent neural network (RNN) trained to determine correction values for the gain and / or offset based on a plurality of training data including the known illumination history of the photon count detector and the corresponding gain and / or offset.

[0021] In an embodiment of the present invention, the artificial intelligence module is a recurrent neural network configured to be trained to determine correction values using a dataset including a plurality of illumination histories of the photon count detector and the corresponding gain and / or offset.

[0022] The recurrent neural network can be trained in this embodiment of the present invention using a sufficient number of training datasets including the irradiation history Φi(E,t), t < 0, and the corresponding offset and gain responses Oi(t), Gi(t), t > 0 of the photon count detector. The illumination history can be sampled before t = 0, and the offset and gain, which are inputs to the training, can include future parts, i.e., values for t > 0. Since the training datasets can be obtained by phantom scans, a known number of photons impinge on the detector at defined time intervals. The determination of the offset and gain can be based on measurements or can be estimated from the appearance of deviations from the nominal detector output, i.e., the output of the detector after a long period of non-illumination with a known spectrum. Once the neural network is trained with a sufficient number of inputs, if a given illumination history (O(t), G(t)) = RNN[Φ(E,t)] is assumed, the offset and gain are predicted using the neural network. This can be used to estimate the pulse height of the signal of the detected photons and thus the count number at all energy bins of the photon detection unit. The entire illumination history of the detector, including detector illuminations such as calibration scans, clinical scans, tube conditioning, etc., can be considered to successfully infer the offset and gain in the polarization state or equivalently.

[0023] In an embodiment of the present invention, the determination unit is configured to determine a correction value based on a physical model of the photon counting detector that takes into account the illumination history of the photon counting detector.

[0024] The correction value can be determined by a physical model that takes into account the space charge or the trapped charge in the sensor as a function of the illumination history, thereby allowing prediction of the correction values for gain and offset. Due to gain drift, the registered count number may change.

[0025] In an embodiment of the present invention, the physical model assumes a constant capture probability of charge carriers generated by photons detected in the photon counting detector and an exponential decay of the charge carriers captured over time.

[0026] The correction value can be determined based on a physical model that uses differential equations to estimate the polarization offset and the decay of the gain. In this embodiment, for example, an empirical model that explains the reasonable physical behavior of the offset and the gain, such as the decay of the polarization offset and the gain to the nominal value after illumination, preferably within a specific time constant of exponential decay, can be explained by the analysis means. The accumulation of polarization and the associated offset and gain may depend on illumination. In this model, differential equations can be used to provide an approach for determining the correction factor. By determining the capture probability and the capture release time constant, a physical model can be described that explains the polarization state of the sensor at any given point in time depending on the illumination history. The capture probability of the charge carriers can be considered to be constant in this embodiment of the present invention, thereby resulting in an increase in the number of trapped charge carriers proportional to the amount of photon flux or energy accumulated in the photon counting detector over a defined time interval. The decay of the capture state can be considered an exponential decay with a specific time constant. However, the model can consider multiple different capture states, each of which decays with a different time constant.

[0027] According to another aspect of the present invention, a computed tomography system including a system according to any of the foregoing embodiments is provided.

[0028] As described in the foregoing embodiments, the computed tomography system can include a system for correcting the number of counts in the energy bin of X-ray photons detected by the photon-counting detector of the spectral computed tomography system in order to provide a medical image with reduced artifacts due to polarization of the detector.

[0029] According to another aspect of the present invention, a method for correcting the number of counts in the energy bin of X-ray photons detected by a photon-counting detector for a spectral computed tomography system is provided. The method includes steps of detecting photons, providing the pulse height of each of the detected photons, and storing the number of counts in the energy bin according to the pulse height of each of the detected photons. The method further includes steps of determining a correction value for the gain and / or an offset of the photon-counting detector based on the illumination history of the photon-counting detector, and correcting the number of counts in the energy bin of the detected photons according to the correction value. The step of correcting the number of counts in the energy bin of the detected photons is performed in the forward model of the spectral computed tomography system by adapting the response function of the photon detection unit according to the correction value for the gain and / or the offset of the photon-counting detector. The space charge effect, polarization effect, and / or deviation of the number of counts in the energy bin due to charge trapping in the photon-counting detector are corrected.

[0030] The method according to the invention corrects the number of counts in the energy bins of the photons detected by a photon counting detector. In a first step, photons are detected and the pulse height of the detected photons is provided. In a second step, the number of counts in the energy bins of the detected photons is stored according to the pulse height of the detected photons. In a third step, a correction value for the gain and / or offset of the photon counting detector based on the illumination history is determined. In a fourth step, the number of counts in the energy bins of the detected photons is corrected according to the correction value for the gain and / or the correction value for the offset of the photon counting detector.

[0031] This method can be applied to the analysis of patient attenuation coefficients and material compositions, which are imaged by a computed tomography system that measures X-ray photons entering different energy bins according to the measured pulse height. Therefore, the patient attenuation coefficient and material composition to be imaged are derived using the number of counts in one or more energy bins, preferably four or five bins. The number of counts can be corrected directly in the energy bins. However, the number of counts can also be corrected by providing a forward model for predicting the value of the number of X-ray photons detected in the energy bins that account for the polarization state of the detector. The forward model can include the relationship between the attenuation coefficient and material composition, the number of counts in the energy bins, and the response function of the photon detection unit. By correcting the pulse height of the photons in the response function with the correction value for the gain and / or offset of the photon counting detector, the material composition can be derived by this indirect correction of the number of counts in the energy bins.

[0032] In an embodiment of the invention, the step of determining the correction of the correction value for the gain and / or the value of the offset of the photon counting detector based on the illumination history is performed by an artificial intelligence module.

[0033] An artificial intelligence module can be used to determine a correction value for gain and / or a correction value for offset. The artificial intelligence module can comprise a recurrent neural network trained with multiple sets of training data, each including an illumination history and a corresponding gain and offset. Alternatively, a set of training data can include a correction factor for gain and / or a correction factor for offset corresponding to an illumination history. Thus, the artificial intelligence module can be trained to determine a correction value for gain and / or a correction factor for offset when any illumination history is provided to the artificial intelligence module.

[0034] In an embodiment of the present invention, the step of determining a correction value for gain and / or a correction value for offset of a photon-counting detector based on an illumination history is performed based on a physical model of the photon-counting detector that takes into account the illumination history of the photon-counting detector.

[0035] The correction value for gain and / or offset can, in this embodiment of the present invention, be determined based on the physical model of the photon-counting detector and the physical processes in the photon-counting detector. The physical model can be based on an illumination history that takes into account the flux of photons previously detected by the photon-counting detector. The physical model can be based on the assumption that the capture probability of charge carriers is constant, thereby resulting in an increase in the captured charge carriers proportional to the photon flux or to the number of free charge carriers generated in the photon-counting detector. The number of captured charge carriers can, in this embodiment of the present invention, potentially decrease in an exponential decay. In this model, one or more capture states with different decay time constants can be assumed. By determining the number of charge carriers captured according to the illumination history, the space charge can be determined to effect a gain reduction or an offset change. Thus, a correction value can be determined for use in correcting the count number in the detected photon energy bin.

[0036] According to another aspect of the present invention, there is provided a computer program element that, when executed on a processing unit, instructs the processing unit to execute the method according to any of the foregoing embodiments.

[0037] The computer program element can be executed on one or more processing units that are instructed to execute a method for correcting the number of counts in the energy bins of X-ray photons detected by a photon-counting detector of a spectral computed tomography system.

[0038] According to another aspect of the present invention, there is provided a processing unit configured to execute the computer program element according to the foregoing aspect of the present invention.

[0039] The processing unit can be distributed over one or more different devices that execute the computer program element according to the present invention.

[0040] Thus, the advantages provided by any of the above aspects apply equally to all other aspects, and vice versa.

[0041] In summary, the present invention relates to a system and method for correcting the number of counts in the energy bins of X-ray photons detected by a photon-counting detector for a spectral computed tomography system. The illumination history of the photon-counting detector is considered to determine the gain and / or offset of the photon-counting detector. The number of counts in the energy bins of the detected photons is corrected according to a correction value corresponding to the determined gain and / or offset.

[0042] The above aspects and embodiments will become apparent from and will be elucidated with reference to the exemplary embodiments described below. Exemplary embodiments of the present invention are described below with reference to the following drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043]

Figure 1

Figure 2

Figure 3

[0044] FIG. 1 is a schematic configuration of a system 100 for correcting the count number 115 in the energy bin of X-ray photons 112 detected by a photon-counting detector 111 of a spectral computed tomography system 300 according to an embodiment of the present invention. Photons 112 incident on the photon-counting detector 111 of the photon detection unit 110 are detected by the photon detection unit 110. The photon detection unit 110 provides the pulse height of the detected photons 112. The storage device 120 is configured to store the count number 115 in the energy bin according to each pulse height 113 of the detected photons 112. The illumination history 125 is provided to the determination unit 130. The determination unit 130 may include an artificial intelligence module 131 in an embodiment of the present invention. The determination unit 130 is configured to determine a correction value 135 for the gain or an offset correction value 135 of the photon-counting detector 111 based on the illumination history 125. The correction coefficient 135 and the count number 115 are provided to the correction unit 140, and the correction unit 140 is configured to correct the count number 115 in the energy bin of the detected photons 112 according to the correction value 135.

[0045] FIG. 2 shows a schematic configuration of a computed tomography system 300 including a system 100 for correcting the count number 115 in the energy bin of X-ray photons 112 detected by a photon-counting detector 111 of a spectral computed tomography system 300 according to an embodiment of the present invention. The computed tomography system 300 further includes a processing unit 200 communicatively connected to the system 100. The processing unit 200 can execute a computer program element that instructs the processing unit 200 to execute the method according to the present invention. Accordingly, the processing unit 200 can control the system 100 according to the present invention. The system 100 can be part of the computed tomography system 300. The photon-counting detector 111 can be a detector of the computed tomography system 300, thereby detecting X-ray photons of the computed tomography system 300.

[0046] FIG. 3 shows a block diagram of a method for correcting the number of counts 115 in the energy bin of X-ray photons 112 detected by a photon counting detector 111 for a spectral computed tomography system 300 according to an embodiment of the present invention. This method has a first step of detecting photons 112 and providing the pulse height 113 of each of the detected photons. After this step, a second step of storing the number of counts 115 in the energy bin according to the pulse height 113 of each of the detected photons 112 follows. In a third step, a correction value 135 for the gain and / or a correction value 135 for the offset of the photon counting detector 111 are determined based on the illumination history 125 of the photon counting detector 111. In a fourth step, the number of counts 115 in the energy bin of the detected photons 112 is corrected according to the correction value 135.

[0047] The physical model for determining the correction value is based on the following assumptions. First, there is no effect of permanent radiation damage. This involves the detector returning to its normal unpolarized state after a sufficiently long non-illuminated time and providing the nominal gain before illumination. Second, the polarization state S(t) determines the output at time t, i.e., the gain and / or offset of the photon counting detector. Third, the polarization state S(t) of the photon counting detector is uniquely determined by the illumination history. Fourth, the polarization state S(t) causes changes in the offset and / or gain of the photon counting detector.

[0048] Assume that the illumination history of the detector is given by a rate spectrum Φ(E,t) having the number of unit photons per energy bin per time interval incident on the detector at time t. Further, S(t) represents the polarization state of the detector at time t. Next, the above assumptions can be converted into mathematical expressions.

[0049] 1) - When -T < t < 0 and T is sufficiently large, if Φ(E,t) = 0, then S(0) = S 0 where S 0indicates the state of the unpolarized detector.

[0050] 2) N b (t) = N b [Φ(E, t), S(t)], where N b (t) is the number of counts measured at energy bin b at time t. This is determined by the incident rate spectrum at time t, Φ(E, t), and the polarization state of the detector at time t, S(t).

[0051] 3) S(t) = S[Φ(E, t’ < t)]. The polarization state of the detector at time t is a function of the illumination history.

[0052] 4) O[S(t)], G[S(t)]. The offset and gain are functions of the polarization state and provide a sufficient description of the polarization state for estimating the expected number of counts at energy bin b.

[0053] The determination of the polarization state S(t) from the illumination history of the photon-counting detector is described in 3). This can be explained using semiconductor physics, including the movement and lifetimes of electrons and holes, the density of impurities, and the physics of metal-semiconductor transitions at the electrodes. There are many physical origins leading to the physical models for the time scales on which these effects occur. However, during these time scales, for an unpolarized sensor, the nominal count rate or the response to individual photons is altered. These deviations in the count rate can be corrected, aiming to obtain a reproducible response of the detector. This functions based on the insight that the output of the detector at a given instant of operation is determined only by the incident spectrum, flux, and illumination history.

[0054] If the offset and / or gain can be uniquely determined from the illumination history, the problem can be considered solved. Equation (1) N b (t) = N b [Φ(E, t), O(t), G(t)], The above equation shows that, when the offset and / or gain are known and the pile-up is understood, the output of the photon-counting detector can be predicted for a given spectral rate incident on the detector. If the offset and gain can be uniquely determined from the illumination history, reliable estimates of the photon counts at each energy bin can be generated for any state of polarization of the detector. Only holes are affected by trapping, and it can be assumed that there is one type of trapping with a trapping release time τ in the semiconductor material. Next, the model of the total trapped positive charge Q(t) can be described as in Equation (2).

Number

[0055] The model of the generation current j(t) can be given, for example, by Equation (3).

Number

[0056] Equation (2) can be solved for any irradiation history j(t). The general solution of the ordinary differential equation can be represented by the inverse Laplace transform L -1 as follows.

Number

Number

[0057] L -1 can be obtained numerically by the Mellin inversion theorem, or using the Fourier space instead of Laplace. Note that the definition of the Laplace transform used here assumes j(t) = 0 for negative t. Since all detector irradiations can be considered to occur at positive t, this does not limit the applicability of the equation in any case.

[0058] Experimental observations indicate that there are several different time constants. This means that there are various types or levels of hole trapping. Assuming no interaction between different types of trapping and neglecting the effect of double hole trapping, the model can be extended as follows. This assumption applies as long as all capture probabilities p i are small compared to 1. In this case, Equation (4) is changed to Equation (6). [Number]

[0059] There is evidence of the existence of several hole trapping levels. This is revealed by various time constants observed in the registered counts and sensor leakage current, assuming the temperature is kept constant. Short-term stability, i.e., Lag, has been shown to have time constants in the range of 60 milliseconds to 120 milliseconds. Helical scans have also been shown to exhibit transient phenomena in the range of seconds. Very long acquisitions also show a time component in the time range. Alternative measurements regarding the settling rate of the dark current also show a component in the time range, which is not correlated with temperature changes. All these observations strongly support the co-occurrence of hole trapping with distinct time constants.

[0060] A model for estimating the charge collected by a detector, described by a polarization state characterized by a total trapped charge Q, for an unpolarized sensor, the charge Q 0 is explained by calculating the reduction of the electric field or voltage in the semiconductor material in the presence of the charge Q compared to the charge Q. This model gives the following.

Equation

[0061] As outlined above, the strongest evidence regarding the origin of the banding artifacts goes back to the change in gain due to polarization. If the polarization also changes the offset of the measured pulse height spectrum, the model for the time-dependent offset O(t) as a function of the polarization state of the sensor, explained by the total trapped charge at time t, is as follows. Equation (8) O(t)=O 0 +ηQ(t), where O 0 is the offset of the unpolarized sensor and η is a constant. Since the offset effect does not seem to be as prominent as the gain effect, it does not seem necessary to include a non-linear term in Q(t). However, a non-linear term can also be considered.

[0062] In an embodiment of the present invention, the number of counts in the energy bin can be explained by Equation (9).

Equation

[0063] Here, R(E,U) is the response function. This is a function that represents the probability that an X-ray photon interacts in a photon-counting detector and the incident energy E is detected at the pulse pulse height U. Considering patient attenuation, Equation 9 can be described as Equation 10.

Equation

[0064] From this equation, the patient attenuation A i can be inferred from the number of counts measured in the unpolarized case. Therefore, A i is derived from the measured number of counts N b . Clearly, if the response changes due to the polarization of the photon detector, this is not useful and A i will be incorrect.

[0065] In the case of a polarization sensor, the response function changes according to Equation 11 below.

Equation

[0066] K(t) and O(t) can be identified with the quantities given by the above equations (7) and (8) with reference to the gain and offset. The above (10) can be regarded as the forward model in the case of a polarization sensor if R(E,U) is replaced by R pol (E,U,t).

[0067] Although the present invention has been illustrated and described in detail in the drawings and the foregoing description, such illustrations and descriptions should be considered to be illustrative or exemplary and not restrictive. The present invention is not limited to the disclosed embodiments. Other modifications to the disclosed embodiments will be understood and achieved by those skilled in the art of practicing the claimed invention from a study of the drawings, the disclosure, and the dependent claims.

[0068] In the claims, the word "comprising" does not exclude other elements or steps, and the singular form does not exclude the plural. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used advantageously. The reference signs in the claims should not be construed as limiting the scope.

Explanation of Reference Signs

[0069] 100 System 110 Photon Detection Unit 111 Photon Counting Detector 112 Photon 113 Pulse Height 115 Count Number 120 Memory Device 125 Lighting History 130 Decision Unit 131 Artificial Intelligence Module 135 Correction Value 140 Correction Unit 200 Processing Unit 300 Computed Tomography System

Claims

1. A system for correcting the number of counts in the energy bins of X-ray photons detected by a photon-counting detector for a spectral computed tomography system, the system comprising: A photon detection unit comprising the photon-counting detector, the photon detection unit detecting photons and providing the pulse height of each of the detected photons; A storage device for storing the number of counts in the energy bins according to the pulse height of each of the detected photons; A determination unit for determining a correction value for the gain and / or an offset correction value of the photon-counting detector based on the illumination history of the photon-counting detector; A correction unit for correcting the deviation of the number of counts in the energy bins caused by the space charge effect, polarization effect, and / or charge trapping in the photon-counting detector, according to the correction value of the gain and / or the correction value of the offset of the photon-counting detector, the number of counts in the energy bins of the detected photons; In a system, The correction unit adapts the response function of the photon detection unit according to the correction value of the gain and / or the correction value of the offset of the photon-counting detector, in the forward model of the spectral computed tomography system, to correct the number of counts in the energy bins of the detected photons. A system characterized by that.

2. The system according to claim 1, wherein the illumination history includes the flux of photons previously detected by the photon-counting detector at time intervals.

3. The system according to claim 1 or 2, wherein the determination unit comprises an artificial intelligence module for determining the correction value of the gain and / or the offset correction value of the photon-counting detector based on the illumination history.

4. The system according to claim 3, wherein the artificial intelligence module uses a data set including a plurality of illumination histories of the photon-counting detector and the corresponding gains and / or offsets to determine the gain of the photon-counting detector. A recurrent neural network trained to determine the correction value and / or the offset correction value.

5. The system according to claim 1 or 2, wherein the determination unit determines the correction value of the gain and / or the correction value of the offset of the photon count detector based on a physical model of the photon count detector, taking into account the illumination history of the photon count detector.

6. The system according to claim 5, wherein the physical model assumes a constant capture probability of charge carriers generated by photons detected by the photon count detector and an exponential decay of charge carriers captured over time.

7. A computed tomography system comprising the system according to claim 1 or 2.

8. A method for correcting the number of counts in an energy bin of X-ray photons detected by a photon count detector for a spectral computed tomography system, the method comprising: detecting photons and providing the pulse height of each of the detected photons; storing the number of counts in the energy bin according to the pulse height of each of the detected photons; determining a correction value of the gain and / or a correction value of the offset of the photon count detector based on the illumination history of the photon count detector; correcting the number of counts in the energy bin of the detected photons according to the correction value of the gain and / or the correction value of the offset of the photon count detector, wherein the space charge effect, polarization effect, and / or deviation of the number of counts in the energy bin due to charge capture in the photon count detector are corrected. The method, wherein the step of correcting the number of counts in the energy bin of the detected photons is performed in the forward model of the spectral computed tomography system by adapting the response function of the photon detection unit according to the correction value of the gain and / or the correction value of the offset of the photon count detector.

9. The method according to claim 8, wherein the step of determining the correction value of the gain and / or the correction value of the offset of the photon count detector based on the illumination history is performed by an artificial intelligence module.

10. The step of determining the correction value of the gain and / or the correction value of the offset of the photon count detector based on the illumination history is performed based on a physical model of the photon count detector that takes into account the illumination history of the photon count detector, according to the method of claim 8.

11. A computer program for correcting the number of counts in the energy bin of X-ray photons detected by a photon count detector for a spectral computed tomography system, the computer program comprising: when executed by a processing unit, to the processing unit: obtaining counts from the storage device for storing the number of counts in the energy bin according to the pulse height of the photons detected by a photon detection unit that detects photons and provides the pulse height of each of the detected photons to the storage device; determining a correction value of the gain and / or a correction value of the offset of the photon count detector based on the illumination history of the photon count detector; correcting the number of counts in the energy bin of the detected photons according to the correction value of the gain and / or the correction value of the offset of the photon count detector, the correction being such that the space charge effect, the polarization effect, and / or the deviation of the number of counts in the energy bin due to charge trapping in the photon count detector are corrected, in a computer program that instructs to perform the steps; The step of correcting the number of counts in the energy bin of the detected photons is performed in the forward model of the spectral computed tomography system by adapting the response function of the photon detection unit according to the correction value of the gain and / or the correction value of the offset of the photon count detector, characterized by the computer program.

12. A processing unit that executes the computer program according to claim 11.

Images