WO2014192935A1

WO2014192935A1 - Photon-counting x-ray computed tomography device and photon-counting x-ray computed tomography method

Info

Publication number: WO2014192935A1
Application number: PCT/JP2014/064490
Authority: WO
Inventors: ユージョウ，; シャオランワン，
Original assignee: 株式会社東芝; 東芝メディカルシステムズ株式会社
Priority date: 2013-05-30
Filing date: 2014-05-30
Publication date: 2014-12-04

Abstract

A photon-counting X-ray computed tomography device according to the present embodiment is provided with: an X-ray tube (101) which generates X-rays; an X-ray detector (103) which detects X-ray photons generated from the X-ray tube (101), and generates output signals corresponding to the number of the detected X-ray photons with respect to each of at least three energy bins; a support mechanism (102) which supports the X-ray tube (101) rotatably about a rotation axis; a combining unit (117) which selects at least two energy bins to be combined on the basis of the numbers of the X-ray photons in the respective energy bins, and combines the numbers of the X-ray photos in the selected energy bins to thereby acquire a combined output signal in a combined energy bin obtained by combining the selected energy bins; and a reconstruction unit (114) which reconstructs an image using the combined output signal.

Description

Photon counting X-ray computed tomography apparatus and photon counting X-ray computed tomography method

Embodiments of the present invention generally relate to spectral computed tomography (CT) and relate to a particular technique of weighting to improve noise in data acquired prior to image reconstruction. The present embodiment relates to a photon counting X-ray computed tomography apparatus and a photon counting X-ray computed tomography method.

Dual energy X-ray CT scan data is obtained at two energy levels. Specifically, the X-ray tube is set to an energy level of low tube voltage and high tube voltage of 80 kilovolts and 120 kilovolts. A dual X-ray source CT scanner is equipped with two X-ray sources, each operating at a different energy level to generate two data sets. On the other hand, in the sandwich detector, the upper layer records low energy data while the lower layer records high energy data. In order to use dual energy data for material classification, the projection data undergoes a preconstruction decomposition.

More generally, spectral information is acquired at two or more energy levels for a particular X-ray CT scanner. Specifically, a predetermined number N of energy thresholds is determined according to an average of the thickness of the material, or a scan for air, and the thickness of the base material is directly calculated based on the N sets associated with the measurement data. The In this regard, all detector units and projection views share the same threshold setting. In practice, it is desirable to change the threshold level between certain views as the spectrum changes.

As the spectrum changes during the scan, the energy band of the energy bin subject to photon counting needs to change according to the situation in order to maintain the low noise level of the obtained data. However, it is theoretically possible to change the threshold dynamically between multiple views, but it is technically challenging because the CT scan is very short. Due to the limitations of current photon counting detector technology, thresholds are not applied correctly in the short time between multiple views as used at one typical rate of 1800 views in 0.5 seconds. It may be. In general, readout electronics, like detectors, have a finite response time and dead time, so considering the currently available technology, the threshold for changing the implementation under the above requirements is limited. .

ボー Since the structure and thickness of the specimen are very uneven, it is difficult to use a bow tie filter. The bowtie filter faces difficulties in matching itself and patient placement, different detection units and multiple views. Moreover, the different units and views depend on significantly changing attenuation and the X-ray spectrum incident on the detector.

Therefore, a universally constant threshold value is a threshold value that can result in undesirable noise balance (noise amount balance) in a huge data set. Unbalanced noise in the resulting data set can cause serious artifacts in the spectral image. In addition to a plurality of energy bins (units of energy bands for performing photon counting), the number of photons in a low energy bin or a high energy bin is more balanced.

For these and other reasons, the prior art described above remains desirable in that it substantially improves the noise balance in images reconstructed from the obtained data with spectral information.

JP 2013-143980 A

An object of the present invention is to provide a photon counting X-ray computed tomography apparatus and a photon counting X-ray computed tomography method capable of balancing noise in energy bins related to image reconstruction.

The photon counting X-ray computed tomography apparatus according to the present embodiment detects an X-ray tube that generates X-rays and X-ray photons generated from the X-ray tube, and according to the detected number of X-ray photons Based on an X-ray detector that generates an output signal for each of at least three energy bins, a support mechanism that rotatably supports the X-ray tube about a rotation axis, and the number of X-ray photons in each of the energy bins, A synthesis that obtains a synthesized output signal in a synthesized energy bin obtained by synthesizing the selected energy bins by selecting at least two energy bins to be synthesized and synthesizing the number of X-ray photons of the selected energy bins. And a reconstruction unit that reconstructs an image using the combined output signal.

FIG. 1 is a diagram showing an embodiment of a multi-slice X-ray CT device or scanner according to this embodiment having a gantry 100 and other devices or units. FIG. 2A is a diagram illustrating an embodiment of a noise balancing device for improving noise balancing according to this embodiment. FIG. 2B is an exemplary set of weight values used in the noise balancing process or device embodiment according to this embodiment. FIG. 3 is a flowchart showing steps or operations having a noise balance improvement process in a computed tomography of a spectrum using a photon counting detector according to the present embodiment. FIG. 4A is a monochrome image reconstructed from noise balance data according to an embodiment of the processing and device according to this embodiment. FIG. 4B is a monochrome image reconstructed from non-noise balanced data according to the present embodiment and processing and device embodiments. FIG. 4C is a diagram showing the image shown in FIG. 4A with lines. FIG. 4D is a diagram showing the image shown in FIG. 4B with lines.

Referring to the figure, the reference numerals clearly indicate the corresponding structure throughout the figure. In particular, FIG. 1 illustrates one embodiment of a multi-slice X-ray CT device, or scanner, according to the current embodiment that includes a gantry 100 and other devices and units. The gantry 100 is illustrated from the front, and includes an X-ray tube 101, an annular frame (support mechanism) 102, and an X-ray detector 103 of a multi-row or two-dimensional array type. The X-ray tube 101 and the X-ray detector 103 are mounted on the annular frame 102 rotating around the axis RA across the subject S in the opposite direction. The rotation unit 107 rotates the annular frame 102 at a high speed, such as 0.4 seconds per rotation, while the subject S moves along the axis RA toward or away from the illustrated page. The support mechanism 102 supports the X-ray tube 101 so as to be rotatable around the rotation axis RA.

The multi-slice X-ray CT device further includes a high voltage generator 109 that supplies a tube voltage to the X-ray tube 101 so that the X-ray tube 101 generates X-rays. In one embodiment, the high voltage generator 109 is mounted on the annular frame 102. The current regulator 118 adjusts the current supplied to the high voltage generator 109 under the control of the system controller 110. X-rays are emitted toward the subject S, and the cross-sectional area of the subject S is indicated by a circle. The X-ray detector 103 is disposed on the opposite side of the X-ray tube 101 across the subject in order to detect the emitted X-rays transmitted through the subject S.

Also with reference to FIG. 1, the X-ray CT device or scanner further comprises a data acquisition device 111 that detects the emitted X-rays and processes the detected signals. In one embodiment, the X-ray detector 103 is implemented using a plurality of photon counting detectors for counting photons in each of a predetermined number of energy bins. For example, the X-ray detector 103 detects X-ray photons generated from the X-ray tube 101 and generates an output signal corresponding to the detected number of X-ray photons for each of at least three energy bins. Each of the plurality of energy bins defines a predetermined range regarding the energy of transmitted X-rays in the X-ray detector 103. After detecting the emitted X-rays with the X-ray detector 103, the data acquisition circuit 104 converts the signal output from the X-ray detector 103 into a voltage signal for each channel, amplifies it, and further converts it to digital. Convert to signal. The X-ray detector 103 and the data acquisition circuit 104 are configured to process a predetermined total number of projections per rotation (Total number of projections per rotation: TPPR).

The data described above is sent to the preprocessing device 106 housed in the console outside the gantry 100 through the non-contact data transmitter 105. The preprocessing device 106 performs specific correction such as sensitivity correction on the raw data. The storage device 112 then stores result data, also called projection data, immediately before the reconstruction process. The storage device 112 is connected to the system controller 110 via the data / control bus together with the reconstruction device (reconstruction unit) 114, the display device 116, the input device 115, and the scan plan support device 200. The scan plan support device 200 has a function of assisting the image technician in order to create a scan plan.

One embodiment of the reconstruction device 114 according to one aspect of the present embodiment is based on a filtered back projection (FBP) technique using noise weights, and an image from projection data stored in the storage device 112. Reconfigure. In the above embodiment, the reconstruction device 114 projects based on a filtered backprojection (FBP) technique using features that emulate a specific iteration result at a predetermined number of iterations according to a predetermined sequential reconstruction algorithm. Reconstruct an image from data. The reconstruction device 114 is implemented by a combination of software and hardware, and is not limited to a specific implementation. In the following description of the reconfiguration device 114, the term “unit” or “device” includes hardware and / or software. In addition, the concept of the reconstruction device 114 can be applied to other modalities including nuclear medicine and magnetic resonance imaging (MRI).

The noise balancing device (synthesizer) 117, in one embodiment, balances (balances) noise in the data obtained by making the number of photon counts substantially uniform among a given number of energy bins. Therefore, it is implemented as software, hardware, or a combination of both. In general, the noise balance device 117 assumes a predetermined number M and a predetermined number of base materials N (N is the number of base materials) in a subject to be image-displayed. Is at least 3 (M> 2) energy bins in each of the plurality of photon counting detectors in the X-ray detector 103 in the CT system, and the number M of energy bins is greater than the number N of base materials (M> N). . The base material is, for example, water, bone, contrast medium or the like. A base material corresponding to a synthetic energy bin, which will be described later, may be set in advance or may be selected according to an operator's instruction. It is theoretically possible that the energy bin and the basis material are the same number (M = N), but the noise balance device 117 is such that the M energy bin is larger than the N basis material to approach the noise balance state. (M> N) is requested. After obtaining the photon count n (E) in each of the M energy bins denoted by E = 1 to M in all of the X-ray detectors 103, the number of photon counts in each of the N energy bins is substantially In order to be equal or as equal as possible or optimal, the noise balancing device 117 combines the M energy bins into N energy bins. That is, the number of photon counts is ideally the same among the N energy bins in the data obtained for each view and the sensing element in each photon counting detector. The noise balancing device 117 then uses a predetermined method, such as a coordinated pursuit (also referred to as Basis pursuit) technique, to find a non-zero L (i) for the photon count of each balanced energy bin. Is related to the thickness L (i) of the material from i = 1 to N. The noise balancing device 117 repeats the above calculation for each of the plurality of detection elements and each of the plurality of views. Finally, an image such as a monochrome image is reconstructed by the reconstruction device 114 according to the material thickness L (i) determined by the noise balancing device 117.

The noise balancing device 117 selects at least two energy bins to be combined based on the number of X-ray photons in each of the M energy bins. The noise balancing device 117 obtains a composite output signal in N synthetic energy bins obtained by synthesizing the selected energy bins by synthesizing the number of X-ray photons of the selected energy bins. At this time, the reconstruction device 114 reconstructs an image using the combined output signal.

The noise balancing device 117 has a small number of X-ray photons when selecting one of the two energy bins adjacent to the predetermined energy bin as the energy bin to be combined with the predetermined energy bin. One energy bin may be selected as a synthesis target. The noise balancing device 117 may acquire two combined output signals respectively corresponding to the two combined energy bins by combining the numbers of X-ray photons belonging to different energy bins. At this time, the reconstruction device 114 reconstructs an image using the two combined output signals.

Also, the noise balancing device 117 may select the energy bin to be synthesized in order to minimize the difference between the two X-ray photon numbers respectively corresponding to the two synthesized energy bins. Further, the noise balance device 117 synthesizes a plurality of weights corresponding to the M energy bins and the N combined energy bins by multiplying the number of X-ray photons of the selected energy bin, thereby combining the N energy bins. A combined output signal in each of the combined energy bins may be acquired.

Also, the noise balancing device 117 may determine the thickness of each of a plurality of base materials equal to the number of composite energy bins (N) using the composite output signal. At this time, the reconstruction device 114 reconstructs an image using the composite output signal corresponding to the determined thickness. Note that the noise balancing device 117 may determine the thickness of each of the plurality of base materials using a predetermined conditional algorithm and the synthesized output signal. In addition, the noise balancing device 117 may determine the thickness of each of the plurality of base materials using a predetermined no-condition algorithm and the synthesized output signal.

It should be noted that the energy bin to be synthesized into the synthesized energy bin may be selected depending on the target substance that the operator pays attention to and the desired scanning condition. The scan conditions are, for example, a tube voltage, a thickness of the subject, and the like. The scan condition is input by the operator via the scan plan support device 200, the input device 115, and the like. The scan conditions may be input by a radiology information management system (RIS) or the like via a network (not shown). The input scan condition is stored in the storage device 112. An attention substance is a substance which an operator pays attention among a plurality of base substances.

That is, the noise balance device synthesizes the energy bins so that when the noise amounts of the combined energy bins are compared, the noise amounts of the two approach each other (so that both noise amounts approach the noise balance state).

In the present embodiment, the decomposition into the base material is performed based on the Latham pair. The pair is spatially and temporally identical in photon counting. Latham is the sum of attenuation coefficients along a ray defined by the point of the X-ray source and the detector element. A Latham pair is two Lathams obtained with two different spectra (eg, high energy and low energy) along the same Latham. Latham pairs are applicable to dual energy sources in one embodiment. As dual energy, for example, a dual energy source that generates X-rays having two energies by switching between high and low tube voltages, and two X-ray tubes by applying two different tube voltages to each of the two X-ray tubes. There is a dual energy source that simultaneously generates two X-rays having different energies from a tube. In other embodiments, three or higher order sets of Latham are applicable to polychromatic X-ray energy CT or spectral CT. In general, the number M of energy bins coincides with the number of laysums generated with the actual spectrum M along the same ray path. A ray path represents a straight line along the passage of an X-ray beam. In order to make M> N and the number of photon counts in each of the N energy bins is substantially equal or as equal as possible or optimal, the noise is synthesized by combining the M energy bins to N It is balanced by making it an energy bin. In the process of synthesis, specific weights are used to make the photon counts substantially equal in N energy bins, with weights from n = 1 to N basis materials and m = 1. To M spectrum. Although it is theoretically possible that the number of energy bins and the number of basis materials are the same number (M = N), the noise balance device 117 is configured so that the number of energy bins M is close to the noise balance state. Requires greater than the number N of base materials (M> N). That is, the processing of the noise balancing method or the embodiment of the noise balancing device 117 does not go into noise balancing in dual energy CT data because at least one extra measurement is required in one ray path.

Next, FIG. 2A is a diagram illustrating that one embodiment of the present noise balancing device 117 approaches a noise balancing state. FIG. 2A has a simple overview in which a set of photon counts are collected across four energy bins from 1 to 4. In other words, at least three collected energy bins are defined according to a fixed threshold for collecting photon counts for a predetermined number of views each having a population of Lathams. A group of laysomes corresponds to, for example, a plurality of laysums included in a cone beam. Spectral data including photon counting sets in the four energy collected energy bins is acquired. In order to balance the photon counts in the collected energy bins, the photon counts are another predetermined two or more processed composites according to a set of weights associated with the base material and energy bins in the subject. Combined (combined) or redistributed among energy bins. In FIG. 2A, the number of energy bins collected is 4 (M = 4), while the number of processed composite energy bins is 2 (N = 2). As illustrated, photon counts are combined (synthesized) into collected low energy bins and high energy bins (composite energy bins) from the collected energy bins.

FIG. 2B is a diagram illustrating a typical set of a plurality of weight values according to the present embodiment as an embodiment of the noise balancing process. Based on the above simple example shown in FIG. 2A, the left side of FIG. 2B lists four typical sets of weight values, each of which is the basis material and energy of the subject. Associated with a specific combination of bins (synthetic energy bins). On the left side of FIG. 2B, the superscript L indicates a low energy bin as a processed composite energy bin, and the subscript number indicates one of the collected energy bins. Similarly, on the right side of FIG. 2B, four typical sets of weight values are listed, each of the four weight values being a specific combination of the basis material and energy bins in the subject (synthetic energy bins). Associated with. On the right side of FIG. 2B, the processed composite energy bin corresponds to a high energy bin. On the right side of FIG. 2B, the superscript H indicates a high energy bin as a processed composite energy bin, and the subscript number indicates one of the collected energy bins.

Continuing with FIG. 2A, the multiple weight values are merely exemplary. A weight value of 1 does not change the normal photon count in a specific energy bin. On the other hand, values less than 1 change the photon count in the associated energy bin. These weight values redistribute the photon counts in the resulting data set to a specific energy bin or multiple energy bins.

Next, FIG. 3 is a flowchart showing an operation or stage having a process for approaching a noise equilibrium state in a computed tomography of a spectrum using the photon counting detector according to the present embodiment. In spectral computed tomography, the process to approximate the noise balance is performed by various methods including software, hardware, and a combination of both. The following steps or actions are optionally performed by the units and devices of the present embodiment, as described above with reference to FIG. 1, but for approaching noise balance in spectral computed tomography using a photon counting detector. Processing is not limited to the performance of this particular embodiment.

Specifically, with reference to FIG. 3, photons are detected or counted for each of a plurality of energy bins at a predetermined photon counting detector or X-ray detector 103 in step S100. The photon counting detector has a predetermined number of detector elements, each detector element having a predetermined number of energy bins (M energy bins) separated by a corresponding number with respect to a fixed threshold at the energy level. . In a typical implementation, the energy threshold level is predetermined and stored in the readout electronic device. There are at least three predetermined numbers M for the collected energy bins to approximate the noise balance. Thus, a first set of photon counts is obtained with at least three energy bins for each of a predetermined number of views having a collection of lathams.

Continuing with FIG. 3, after the photon count is obtained with at least three energy bins (M = 3 energy bins), the photon count is the photons contained in the processed composite energy bin in step S110 according to this embodiment. Synthesized into a second set of counts. The number N of synthetic energy bins is two or more and is equal to the number of base substances in the subject. The number N of composite energy bins is smaller than a predetermined number M of energy bins (N <M) in order to bring the noise in the spectral data closer to equilibrium between the composite energy bins. In approximating the noise balance, a predetermined set of weights is used, each of the predetermined weights uniquely being one of the corresponding principal components (or composite energy bins) and corresponding Is associated with one of the energy bins.

After the noise balance is performed in step S110, the photon count in the composite energy bin is related to the material thickness for each of the base materials in step S120. The photon counts in each of the combined energy bins are now more balanced in noise than before step S110. In step S120, the more noise balanced photon count in each of the combined energy bins is now associated with the material thickness L (i) where i is 1 to M. Non-zero L (i) is determined using a predetermined coordinated perchue technique.

The predetermined coordinate pursuit technique is, for example, basis pursuit, and in the case of this embodiment, the following conditional expression:

The following formula

Is an algorithm for solving
In step S130, it is determined whether steps S100, S110, and S120 are further performed for any remaining elements relating to detector elements or photon counting detectors for every view. If, in step S130, it is determined that steps S100, S110, and S120 have not been completed for each detection element or photon counting detector for every view, the process proceeds with the plurality of steps being either detection elements or photons. Return to step S100 to be performed on the counting detector or the rest in the view. On the other hand, if it is determined in step S130 that steps S100, S110, and S120 are complete for each detector or photon counting detector for every view, the process proceeds with the material thickness of each base material. Proceed to step S140 to reconstruct the image from the noise-balanced photon count in the processed composite energy bin.

Assuming that two basis materials were used, at least three energy bins are involved because the third energy bin is involved in balancing noise between the first and second bins, such as the high energy bin and the low energy bin. A bin (M = 3) is required as an energy bin. For example, when the ray path is long and most of the low energy photons are absorbed by the subject, the count in the third or medium energy bin is added to the low energy bin. On the other hand, for short ray paths, the count in the medium energy bin is added to the high energy bin.

In general, assuming that the number N of base materials is arbitrary, the number M of energy bins is greater than the number N of synthetic energy bins processed (M> N). In other words, the photon counts in the collected energy bins are combined into processed composite energy bins to approximate the noise balance in the energy bins. If the collected energy bin M is equal in number to the processed composite energy bin N, the same number of energy bins will fail to approach the noise equilibrium. At this time, the projection data g _m measured in the energy bin m is expressed by g _m ^(BH) as a radiation effect term, an average attenuation coefficient with respect to the base material n and the energy spectrum m.

As

Is related to the estimated thickness L _n of the base material. The base material is indexed by n to indicate a specific base material between 1 and N. Similarly, energy bins are indexed by specifying specific energy bins between 1 and M energy bins.

Focusing on a particular ray path, there may be only one material, such as soft tissue, and the thickness of the other base material is zero. Therefore, the thickness L _n of the base material is determined by optimizing or minimizing the following evaluation function (L).

At this time, the optimization depends on data conditions.

Here, σ _m represents the noise of the measured projection g _m .

Alternatively, the noise balancing step is arbitrarily skipped and L _n is determined directly by minimizing the evaluation function with additional constraints. In the conditional algorithm, the thickness L of the base material is expressed by Equation (1) as follows:

Is determined from the projection data g based on the inverse function of.

Is a weighted average attenuation coefficient defined by (17) described later.
Subsequently, the evaluation function is minimized to reduce or balance the noise, as represented by equation (2), using a constant a to determine the noise removal step size. .

At this time, the noise is confirmed by assuming the following data conditions, as represented by Expression (3). The noise balance device 117 calculates the basis material from the sum of the projection data g _m measured in the energy bin and the radiation hardening projection data g _m ^(BH) (L) depending on the thickness (L) of the basis material in the energy bin. Mean attenuation coefficient defined by the thickness L, the base material type n and the energy bin m

The weight is determined on condition that the absolute value of the difference obtained by subtracting the sum over the product type n is smaller than the noise σ _m related to the projection data (formula (3)).

Projection data g _m, as represented by the formula (4), by using a radiation quality curing section depends on the thickness of the base material g ^{(BH) (L)} are updated.

The above steps in the conditional algorithm as summarized through equations (1) through (4) are repeated for every view. The conditional algorithm in the above steps is executed such that the non-zero thickness variance is reduced in the base material.

In the unconditional algorithm, the evaluation function Λ (L) is expressed by Equation (5) using β, which is a positive constant that determines the weight of the loss term.

The thickness L of the base material is

Is initialized based on the projection data g.

Subsequently, the slope of the evaluation function Λ (L) is expressed by Expression (6).

Equation (6) is approximated by Equation (7) as follows. In the approximation from Equation (6) to Equation (7), the partial differentiation of g _m ^(BH) (L) by L _n is relatively small compared to the other terms in Equation (6) and can be ignored.

The thickness of the base material is updated by equation (8).

Here, _L ⁿ in the above formula (8) ⁽⁰⁾ shows the current value in the repetition. In addition, L _n in the above formula (8) indicates a value updated in repetition. L ⁽⁰⁾ is a vector having L _n ⁽⁰⁾ as an element.
here,

Or

It is. Finally, the noise is confirmed by assuming the following data conditions as represented by Equation (9).

The above steps in the unconditional algorithm as summarized through equations (6) through (9) are repeated for every view. The above steps are performed so that the non-zero thickness dispersion is reduced in the base material.

In the exemplary processing related to noise reduction according to the present embodiment, a specific evaluation function is used. However, processing for approaching the noise equilibrium state is limited to the evaluation function as a specific example according to the present embodiment. It is not something. The second evaluation function Λ is defined by Equation (10), with the noise for the energy bin m as σ _m ² for a given measurement projection g _m . That is, the noise balancing device 117 uses the evaluation function Λ as the projection data g _m measured in the energy bin and the line hardening hardening projection data g _m ^(BH) (L) depending on the thickness (L) of the base material in the energy bin. Mean attenuation coefficient defined by the base material thickness L, the base material type n, and the energy bin m

The square of the difference obtained by subtracting the sum over the product type n is defined as the sum over the energy bin m of the quotient obtained by dividing the square of the noise σ _m ² for the projection data.

In order to find the minimum value of the evaluation function Λ or to optimize the evaluation function Λ, the dependence of the quality hardening term g _m ^BH on L _n is ignored based on the weight calculation. Therefore, the approximated result takes the same value range as n = 1 to N, and is expressed by the following equation (11) using n ′ = 1 to N as the second index for the base material. The

n '= 1 to N is used to easily distinguish the second use of the index.

The weight value w is defined to perform noise reduction and is determined for each corresponding basis material n and energy bin m and is now normalized to a particular basis material as in equation (12) below. is defined by using the k _n is a factor.

Normalization factor k _n are defined as the following equation (13).

Thus, the measured weighted version of the projection and the thickness of the basis material should have the following relationship, as represented by equation (14):

here,

Is the weighted measurement projection data for the base material n based on all measured spectral energy levels, as defined by equation (15) below.

Similarly,

Is a weighted line hardening term for the base material n based on all measured spectral energy levels, as defined by equation (16) below.

Finally,

Is the weighted average attenuation coefficient for the base material n based on all measured spectral energy levels, as defined by equation (17) below.

In addition to the above two evaluation functions, the third evaluation function is defined to be used together with the process for approaching the noise balance state in the energy bin according to the present embodiment. In order to define the third evaluation function, the following equation (18) is assumed:

Here, the measured projection data g _m is measured by the energy bins m, associated with the estimation of the thickness L _n. The base material g _m ^(BH) is a linear hardening term,

Is the average attenuation coefficient for the base material n and the energy bin m. The basis material is indexed by n to define a particular basis material between 1 and N basis materials. Similarly, energy bins are indexed to define a particular energy bin between 1 and M energy spectra.

Further, the weighted version of the measured projection and the base material thickness as represented by equation (14) is the weight as represented by equations (15), (16) and (17). Is applicable to the relationship represented by the equation (18). By ignoring the linear hardening term in Equation (14) relating to noise estimation for the base material, the following approximation is obtained as in Equation (19).

Like the following formula,

A variable Ξ that is the reciprocal of is defined.

Thus, Expression (19) becomes Expression (21) indicating the thickness of the base material n based on Equations (20) and (15).

Like the following formula,

A weighted version of the variable that is the inverse of

Is defined.

The dispersion covariance regarding the thickness of the base material is now determined by the following equation (23).

Finally, the evaluation function Λ as the weighted noise of the thickness of the base material is expressed by Expression (24). That is, the noise balancing device 117 calculates the evaluation function Λ by the inverse of the weighted average attenuation coefficient for the base material.

And the variance of the base material thickness L defined by the product of the noise squared to the energy bin m σ _m ² .

In order to find the optimal weight as shown below, the above equation need not be solved analytically. A numerical method is used to find the optimal weight.

Referring now to FIGS. 4A and 4B, a pair of images shows some effects of noise balance according to embodiments of the present apparatus and processing. FIG. 4A is a monochrome image reconstructed from noise balanced data according to an embodiment of the present apparatus and processing. FIG. 4B is a monochrome image reconstructed from data that has not been brought close to the noise balance according to the apparatus and processing embodiment of the present invention. Unbalanced noise can cause severe artifacts in monochrome images, as shown in FIG. 4B, and that noise will likely reduce diagnostic capabilities. For this reason, the improved noise nature probably increases the diagnostic ability based on the final image quality. FIG. 4C is a diagram showing FIG. 4A as a diagram. FIG. 4D is a diagram showing FIG. 4B as a diagram.

According to the configuration described above, the following effects can be obtained.
According to the photon counting X-ray computed tomography apparatus according to the present embodiment, an energy bin to be synthesized is selected based on the photon count collected by each of a plurality of energy bins, and the X-rays of the selected energy bins are selected. The number of photons can be synthesized. This energy bin selection can be performed to balance the photon counts belonging to the composite energy bin. Also, the energy bins may be selected such that the dispersion of the base material thickness is small. Thereby, according to this embodiment, the photon count which belongs to a synthetic energy bin becomes substantially the same or close. Therefore, according to the photon counting X-ray computed tomography apparatus according to the present embodiment, it is possible to prevent a specific energy bin from being excessively noisy.

Further, according to the present embodiment, it is also possible to select the energy bin to be synthesized according to the scanning condition or the target substance that the operator pays attention to. Thereby, according to this embodiment, noise is balanced and the image corresponding to the scanning condition or the target substance can be reconstructed.

From the above, according to the present embodiment, the noise in the plurality of synthetic energy bins can be balanced by making the photon counts in the plurality of synthetic energy bins substantially uniform. Thus, as is apparent from FIGS. 4D and 4C, according to the present embodiment, the reconstructed image quality is improved due to noise balance.

Note that the functions according to the embodiment can also be realized by installing a program (medical image reconstruction program) for executing the processing in a computer such as a workstation and developing the program on a memory. At this time, a program capable of causing the computer to execute the technique is stored in a storage medium such as a magnetic disk (floppy (registered trademark) disk, hard disk, etc.), an optical disk (CD-ROM, DVD, etc.), or a semiconductor memory. It can also be distributed.

Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 100 ... Gantry, 101 ... X-ray tube, 102 ... Annular frame (support mechanism), 103 ... X-ray detector, 104 ... Data acquisition circuit, 105 ... Non-contact data transmitter, 106 ... Pre-processing device, 107 ... Rotation unit , 108 ... slip ring, 109 ... high voltage generator, 110 ... system controller, 111 ... data collection device, 112 ... storage device, 114 ... reconstruction device, 115 ... input device, 116 ... display device, 117 ... noise balance device (Combining unit), 118 ... current regulator, 200 ... scan plan support device.

Claims

An X-ray tube that generates X-rays;
An X-ray detector that detects X-ray photons generated from the X-ray tube and generates an output signal corresponding to the detected number of X-ray photons for each of at least three energy bins;
A support mechanism for rotatably supporting the X-ray tube around a rotation axis;
Based on the number of X-ray photons in each of the energy bins, at least two energy bins to be combined are selected, and the selected energy bins are synthesized by combining the number of X-ray photons of the selected energy bins. A combining unit for obtaining a combined output signal in the combined combined energy bin;
A reconstruction unit that reconstructs an image using the combined output signal;
A photon counting X-ray computed tomography apparatus comprising:
When the combining unit selects one of the two energy bins adjacent to the predetermined energy bin as the energy bin to be combined to be combined with the predetermined energy bin, the number of X-ray photons is small. 2. The photon counting X-ray computed tomography apparatus according to claim 1, wherein one of the energy bins is selected as a synthesis target.
The combining unit acquires the two combined output signals respectively corresponding to the two combined energy bins by combining the X-ray photon numbers belonging to the different energy bins.
The photon counting X-ray computed tomography apparatus according to claim 1, wherein the reconstruction unit reconstructs the image using the two combined output signals.
The synthesis unit is
4. Photon counting X-ray computed tomography according to claim 3, wherein the energy bins to be combined are selected in order to minimize the difference between the two X-ray photon numbers respectively corresponding to the two combined energy bins. apparatus.
The combining unit combines a plurality of weights corresponding to the energy bin and the combined energy bin by multiplying the number of X-ray photons of the selected energy bin, thereby combining the combined output signal in the combined energy bin The photon counting X-ray computed tomography apparatus according to claim 1, wherein:
The synthesis unit determines a thickness of each of a plurality of basis materials equal to the number of the synthesized energy bins using the synthesized output signal,
The photon counting X-ray computed tomography apparatus according to claim 5, wherein the reconstruction unit reconstructs the image using the combined output signal corresponding to the thickness.
The photon counting X-ray computed tomography apparatus according to claim 6, wherein the synthesis unit determines the thickness using a predetermined conditional algorithm and the synthesized output signal.
The photon counting X-ray computed tomography apparatus according to claim 6, wherein the combining unit determines the thickness using a predetermined no-condition algorithm and the combined output signal.
The X-ray tube generates two types of X-rays having two different energies,
The X-ray detector generates an output signal corresponding to the number of X-ray photons corresponding to the two types of X-rays for each of at least three energy bins,
The combining unit selects at least two energy bins to be combined based on the number of X-ray photons in each of the energy bins respectively corresponding to the two types of X-rays, and X-rays of the selected energy bins The photon counting X-ray computed tomography apparatus according to claim 1, wherein a combined output signal in a combined energy bin obtained by combining the selected energy bins is acquired by combining the number of photons.
The X-ray tube generates multicolor X-rays having different energy,
The X-ray detector generates, for each of at least three energy bins, an output signal corresponding to the number of X-ray photons corresponding to the energy of the polychromatic X-ray;
The combining unit selects at least two energy bins to be combined based on the number of X-ray photons in each of the energy bins respectively corresponding to the energy of the polychromatic X-ray, and the X of the selected energy bin The photon counting X-ray computed tomography apparatus according to claim 1, wherein a combined output signal in a combined energy bin obtained by combining the selected energy bins is acquired by combining the number of line photons.
The photon counting X-ray computed tomography apparatus according to claim 6, wherein the synthesis unit determines the weight by specifying the non-zero thickness in the base material by minimizing an evaluation function.
The combining unit calculates the thickness of the base material, the type of the base material, and the sum of the projection data measured in the energy bin and the radiation hardening projection data depending on the thickness of the base material in the energy bin. 12. The weight is determined on the condition that an absolute value of a difference obtained by subtracting a sum total over the types of products with an average attenuation coefficient defined by an energy bin is smaller than noise related to the projection data. The photon counting X-ray computed tomography apparatus as described.
The photon counting X-ray computed tomography apparatus according to claim 11, wherein the evaluation function is a sum of thicknesses of the base materials.
The evaluation function is calculated based on the sum of the projection data measured in the energy bin and the radiation hardening projection data depending on the thickness of the base material in the energy bin, and the thickness of the base material, the type of the base material, and the Defined by the sum over the energy bin of the quotient obtained by dividing the square of the difference obtained by subtracting the sum over the type of product with the average attenuation coefficient defined by the energy bin by the square of the noise for the projection data. The photon counting X-ray computed tomography apparatus according to claim 11.
12. The photon according to claim 11, wherein the evaluation function is a variance covariance of the thickness of the base material defined by a product of an inverse of a weighted average attenuation coefficient for the base material and a square of noise for the energy bin. Counting X-ray computed tomography system.
X-rays are generated by the X-ray tube,
Detecting X-ray photons generated from the X-ray tube, and generating an output signal corresponding to the detected number of X-ray photons for each of at least three energy bins;
Based on the number of X-ray photons in each energy bin, select at least two energy bins to be combined,
By synthesizing the number of X-ray photons of the selected energy bin, a combined output signal in the combined energy bin obtained by combining the selected energy bin is obtained,
Reconstructing an image using the composite output signal;
A photon counting X-ray computed tomography method comprising: