Method and apparatus for estimating thickness of material in radiological projection image

Abstract

The invention relates to a method for estimating the thickness of a material in a radiological projection image, comprising the following steps: providing a number N of a plurality of projection images at different recording energies, performing a first decomposition of the N projection images into N thickness maps, performing a second decomposition of a number of main thickness maps based on the N thickness maps and / or based on the corresponding measurements and a number of projection images of the N projection images into N resultant thickness maps, the N thickness maps each representing a thickness of N regions having respectively different materials,-performing a second decomposition of a number of main thickness maps based on the N thickness maps and / or based on the corresponding measurements and a number of projection images of the N projection images, the N result thickness maps respectively represent thicknesses of N regions having respectively different materials. In addition, the invention includes an apparatus and an image forming apparatus.

Description

Technical Field

[0001] The present invention relates to a method and a device for estimating material thickness in a radiographic projection image and an imaging device. In particular, the present invention relates to a projection-based multi-energy decomposition. Background Art

[0002] In spectral imaging, the so-called "dual energy method" is often used in order to visualize, for example, iodine concentrations or bone concentrations. With two exposures at different energies, values ​​for two different materials or thicknesses of different areas can be estimated. For example, in mammography, the thickness of the breast is assumed to be constant, which can be used to separate glandular tissue and iodine concentration.

[0003] However, after compression of the breast, there are often different areas of different thickness, such as the edge of the chest due to the tilt of the paddle from the chest wall to the standing side, skin folds in the area of ​​the armpit, or the breast itself. If the iodine concentration is in such an area, it is not possible to distinguish whether it is a thickness change or a real iodine concentration. This is the problem.

[0004] Furthermore, in X-ray radiography, the aim is to separate bone from water. However, the presence of fat or other material results in an incomplete decomposition and leaves bone artifacts in the "water" image.

[0005] Currently, contrast-enhanced mammography is performed with low-energy and high-energy weighted logarithmic subtraction algorithms for visualizing iodine concentration. This leads to the aforementioned problem that regions of different thicknesses cause non-uniform background intensity after logarithmic subtraction. Several post-processing algorithms attempt to compensate for this non-uniformity while maintaining the iodine concentration in the region.

[0006] In X-ray radiography the principle is the same, but an incomplete subtraction results in bone structures being visible in the "water" image. Currently no correction is applied for this. Summary of the invention

[0007] The object of the present invention is to provide a method and a device for estimating material thickness in radiological projection images as well as an imaging device with which the above-mentioned disadvantages can be avoided.

[0008] This object is achieved by a method according to the invention for estimating material thickness in radiological projection images, a device according to the invention for estimating material thickness in radiological projection images and an imaging device according to the invention.

[0009] The method according to the invention is used to estimate the material thickness in a radiological projection image. The method comprises the following steps:

[0010] - providing (in particular recording or loading from a storage unit) a number N of projection images at respectively different recording energies,

[0011] - performing a first decomposition of the N projection images into N thickness maps, the N thickness maps respectively representing the thickness of N regions of respectively different materials,

[0012] - performing a second decomposition based on N thickness maps and / or on a certain number of main thickness maps based on corresponding measurements and a certain number of projection images among the N projection images into N resulting thickness maps, wherein the N resulting thickness maps respectively represent the thickness of N regions having respectively different materials.

[0013] The provision of projection images is sufficiently known. For example, N projection images can be recorded with a dual-energy or multi-energy X-ray device and resolved directly with the method. However, the projection images can have been recorded earlier and then stored. Then, they are ready to be downloaded from the data storage. The projection images can have been recorded simultaneously or successively. It is only important that the projection images show the same subject in the same position. It is also theoretically possible to correct motion artifacts, for example by image registration.

[0014] Basically, it is only important for the method that two or more projection images are available, which show the same scene at different energies. Even though the method is already advantageous at two different energies, the advantages emerge even more clearly at three or more energies. In order to better understand the method, it can be assumed that three projection images are recorded: H projection image P at the highest recording energy H , M projection image P at the intermediate shooting energy M And the L projection image P at the lowest shooting energy L .

[0015] Advantageously, the recording energy is selected such that the projected images do not overlap in energy (i.e. the same energy is imaged in both images) and the energy is in a range in which at least one material differs from the other materials in terms of its attenuation (hereby always referring to the attenuation of the X-ray radiation). With regard to contrast agents, for example, the recording energy can be selected such that it is in the K-edge of the contrast agent.

[0016] In the process of the invention, not only one decomposition but two decompositions are performed, wherein N thickness maps are created from the projection images (and in the case of a second decomposition also from the main thickness map). In this case, the thickness maps each represent the thickness of N regions of respectively different materials. The thickness maps essentially correspond to the projection images, with the difference that the "image points" of the thickness maps do not display the attenuation in the form of gray tones, but rather the thickness information, for example, in millimeters. The thickness maps usually also indicate the total thickness of the breast at the corresponding "image point".

[0017] In the first decomposition, the N projection images are decomposed into N thickness maps. For this purpose, for example, the P vectors in the N projection images can be calculated into N thickness maps using an N×N matrix. It should be noted that the projection image is formed by image points. Here, each point at the position (x, y) of the thickness map is formed by the corresponding image point at the position (x, y) of the projection image. For example, using the matrix M* and the P vectors (P1(x, y), P2(x, y), ..., P at the corresponding coordinates (x, y) N (x, y)) can be calculated point by point as d vectors (d1(x, y), d2(x, y), ..., d N (x, y)), where d = M*·P. The decomposition will be described in more detail below.

[0018] The second decomposition is performed after the first decomposition. Unlike the first decomposition, now not only the projection map but also at least one main thickness map D is included in the calculation. For example, the vector (D, P1+P2, P3, ..., PN) is formed as an input vector and multiplied with the matrix K*.

[0019] The matrices M* and K* can be easily predefined. Preferably, the matrices M* and K* are calculated as inverse matrices of M and K, which are formed by attenuation coefficients for the relevant materials. This will be explained in more detail below.

[0020] The main thickness map D can be directly based on the N thickness maps, ie, for example, D = d1 + d2 + ... + d N , but the main thickness map can also be based on corresponding measurements, for example by measuring the imaged breast directly or based on a measured model. It should be noted here that D essentially represents the total thickness of the breast. Thus, D(x, y) is a map that represents the total thickness of the breast at the position (x, y).

[0021] It is now possible to simply add the thickness map to the N projection images and to calculate the (N+1)xN vector as input vector using the Nx(N+1) matrix. However, as is also shown below, it is advantageous to calculate the matrix (K*) which is the inverse of the matrix K in which the attenuation coefficients are used. A square matrix is ​​thus provided. For this purpose, one projection image can be omitted from the N projection images for each main thickness map, for example a projection image which only provides a small contribution. However, in order to be able to work with as much information as possible, it is advantageous to combine two or more projection images. This will be described in more detail below.

[0022] The resulting thickness map can then be output, for example directly on a display for evaluation or it can be stored in a data memory.

[0023] The device according to the invention is used for estimating material thickness in radiological projection images. The device is preferably designed to carry out the method according to the invention and comprises the following components:

[0024] - a data interface for receiving a plurality of projection images, N in number, at respectively different recording energies, preferably an H projection image captured at a highest recording energy (e.g., among three recording energies), an M projection image captured at an intermediate recording energy (e.g., among three recording energies), and an L projection image captured at a lowest recording energy (e.g., among three recording energies),

[0025] - a decomposition unit designed to:

[0026] i) a first decomposition of the N projection images into N thickness maps, the N thickness maps respectively representing the thickness of N regions of respectively different materials, and

[0027] ii) a second decomposition based on the N thickness maps and / or on a number of main thickness maps based on corresponding measurements and a decomposition of a number of the N projection images into N result thickness maps, the N result thickness maps respectively representing the thickness of N regions of respectively different materials,

[0028] Preferably, the three projection images are decomposed into three thickness maps, wherein the three thickness maps respectively represent the thicknesses of three different materials.

[0029] The functions of the components of the device have been described in the course of the method.

[0030] The imaging device according to the present invention is preferably a multi-layer detector or a photon counting detector. The imaging device is designed for spectral imaging of an object by means of radiation and comprises: a radiation source, which is designed to emit a spectrum including N different imaging energies; a detector unit, which is designed to detect the N imaging energies; and a device according to the present invention.

[0031] There are several possibilities for recording multiple energies. The radiation source can emit a broadband spectrum, in which all desired recording energies are contained. However, the radiation source can also emit individual (narrowband) spectra for each recording energy or for subgroups of recording energies (simultaneously or sequentially).

[0032] There are also multiple alternatives for the detector. For example, a single detector can be used, which is sensitive to all recording energies. In the case of broadband radiation sources, a filter should be used, which allows only one recording energy to pass in order to record the projection image in each case. Alternatively, the radiation source can emit a continuous narrowband spectrum for each recording energy.

[0033] In another example, an energy selective detector can be used, or a plurality of detectors can be used, each of which is active only in a narrow energy band. In this case, a plurality of acquisitions can be carried out simultaneously at different acquisition energies.

[0034] The present invention can be implemented in particular in the form of a computer unit with suitable software. For this purpose, the computer unit can have, for example, one or more cooperating microprocessors or the like. In particular, the present invention can be implemented in the form of suitable software program parts in a computer unit. The implementation as far as possible according to software has the advantage that the computer units that have been used so far can also be retrofitted in a simple manner by software or firmware updates in order to work in the manner according to the present invention. In this regard, the object is also achieved by a corresponding computer program product with a computer program, which can be directly loaded into the memory device of the computer unit, and the computer program has program segments so that when the program is executed in the computer unit, all steps of the method according to the present invention are performed. In addition to the computer program, such a computer program product can include additional components, such as documents and / or additional components, and also hardware components, such as hardware keys (dongles, etc.) for using the software.

[0035] A computer-readable medium, such as a memory stick, a hard disk or other removable or fixedly installed data carrier, can be used for transmission to a computer unit and / or storage at or in a computer unit, on which program segments of a computer program that can be read in and executed by a computer unit are stored.

[0036] Other particularly advantageous designs and improvements of the invention are derived from the dependent claims and the following description, wherein claims of a claim category can also be improved into another claim category similarly to the claims and the description, and in particular individual features of different embodiments or variants can also be combined into new embodiments or variants.

[0037] Preferably, the decomposition of the projection image P into the thickness map d is performed using a preset N×N matrix. This has been explained above. For example, for the first decomposition, the P vector from the projection image can be multiplied with a suitable N×N matrix M*, or for the second decomposition, the PD vector formed by the projection image and at least one main thickness map can be multiplied with a suitable N×N matrix K*. Preferably, the matrix (for the first decomposition and / or the second decomposition) is composed of the attenuation coefficients μ of the material for different energies. ij The inverse matrix of the matrix formed.

[0038] In a preferred case, in which recording is performed using (at least) three recording energies,

[0039] - providing an H projection image, which has been recorded at the highest recording energy,

[0040] - providing an M-projection image, which has been acquired at an intermediate acquisition energy, and

[0041] - Providing an L projection image, which has been captured at the lowest capture energy.

[0042] The three projection images are then decomposed into three (or more if necessary) thickness maps in a first decomposition and together with at least one main thickness map in a second decomposition. Figure 1 The thickness map is also decomposed into three (or more if necessary) thickness maps, namely the result thickness maps.

[0043] This is explained in more detail below using a preferred example in which H projection images P are recorded with three different recording energies. H , M projected image P M and L projected image P L The example described can be easily extended to N energies or reduced to two energies using the principles shown.

[0044] For different materials, the corresponding attenuation coefficient μ is known, as will be described below with respect to Figure 2 Therefore, the attenuation coefficients μ for the three materials at energies H, M and L are i (i=1, 2, 3) are known and can be expressed as coefficients of the matrix M:

[0045]

[0046] For the P vector (P H , P M , P L ) and the d vector (d1, d2, d3) apply: P = M·d, i.e. the corresponding thickness multiplied by the appropriate attenuation coefficient in the matrix yields the expected image value of the projection image. If the thickness of the corresponding material is now completely known (this would be a thickness map), the projection image can be calculated.

[0047] However, since the projection images are known and the thickness map is to be determined, it is advantageous to transform the matrix M into a matrix M*. The coefficients m then follow ij , the coefficient m ij It can be calculated from the inverse matrix of the matrix M in (1) above:

[0048]

[0049] Therefore, it is then possible to project the image P from H H , M projected image P M and L projected image P L Three thickness maps d1, d2, and d3 are calculated, that is, the projection image is decomposed into thickness maps.

[0050] As described, further projection images recorded with different energies can also be provided in addition. Each different energy can create a thickness map for a different material or improve previous results.

[0051] In the second decomposition, the process is slightly different, since at least one main thickness map is used there. In the following example, it is assumed that a main thickness map D=d1+d2+d3 has been previously calculated and formed. It is also possible to carry out a measurement of the thickness D(x, y) during the recording, or to derive the main thickness map from a model. However, the calculation of D described at the beginning is particularly advantageous, since no information different from the projection images has to be provided. Since three coefficients should also be used, the two projection images, namely the M projection image and the L projection image, are combined into one ML projection image P ML . Using the matrix:

[0052]

[0053] For the PD vector (D, P ML , P H ) and the final df vector (df1, df2, df3) apply: PD = K · df. Here, in order to calculate the final thickness map, the matrix K should now also be transformed into the matrix K*. Then it follows that the coefficient k ij , coefficient k ijFrom the inverse matrix of matrix K in (3) above, we can calculate:

[0054]

[0055] Preferably, in one embodiment of the method, one of the thickness maps is

[0056] - a K thickness map, which represents the thickness of the contrast agent area, and / or

[0057] - a G thickness map representing the thickness of the glandular tissue, and / or

[0058] - an A thickness map, the A thickness map representing the thickness of the adipose tissue, and / or

[0059] - a B thickness map, said B thickness map representing bone thickness, and / or

[0060] - A W thickness map, which represents the thickness of the area containing water.

[0061] Preferably, in particular for a method with three acquisition energies, one of the three thickness maps is a K thickness map, another thickness map is a G thickness map, and another thickness map is an A thickness map.

[0062] Preferably, the recording energy is based on the attenuation properties of the materials on which the thickness map is based. As described above, the recording energy should be selected from an energy range in which the attenuation coefficient of at least one of the materials differs significantly from the attenuation coefficients of the other materials. Preferably, the radiation energy is selected at which the attenuation properties of all materials or of at least one material differ maximally relative to the respective other materials. Preferably, the recording energy is selected (when recording the projection image) such that it is located in the descending portion of the absorption edge of one of the materials.

[0063] The lowest recording energy is preferably between 10 keV and 100 keV. The intermediate recording energy is preferably between 20 keV and 130 keV. The highest recording energy is preferably between 30 keV and 150 keV. However, the recording energy is highly dependent on the material, according to which the material is to be separated. It should be noted that generally an energy is selected that achieves a good energy separation of the desired material.

[0064] According to a preferred method, the first decomposition of the projection image into thickness maps is performed multiple times, and the thickness maps are used to correct the beam hardening of the decomposed projection image and / or the coefficients of the matrix used in the decomposition. The first decomposition and correction are performed multiple times before the second decomposition, which only needs to be performed once, but the corresponding correction can also be performed multiple times.

[0065] It should be noted that work is already done at different acquisition energies and that different materials absorb differently at different energies. This is compensated for by a correction for beam hardening. Basically, the (pseudo) single-energy attenuation is calculated from the multi-energy attenuation.

[0066] In this case, a polynomial correction is preferably applied. In this case, the corrected projection image P korr (x, y) can be obtained from the projection image P(x, y) by means of the coefficients c which can be determined or calculated by test measurements for each image point (x, y) a The correction is performed according to the following formula with the previously determined order A of the polynomial:

[0067]

[0068] That is, in principle the polynomial c0+c1P+c2P 2 +......+c N P N It should be noted that the coefficient c also depends essentially on the thickness map, ie on the thickness of the material through which the ray must pass before it hits the image point of the detector. It therefore applies that for each image point, c a Basically it depends on (d1, d2, ..., d N ). Here, the correlation can be completely ignored for some materials. If, for example, glandular, fatty and iodine thickness maps are generated, it has been shown that the correlation with the main thickness map D (three thickness maps added) and the glandular thickness map G, if necessary together with the iodine thickness map, is sufficient. Therefore, it is preferably assumed that c a (d g / D, D) or c a (d j / D,d g / D, D). This can be used for different g ,d j and D are calculated or measured.

[0069] It should be noted that for the image recorded at the beginning of the method ("initial image"), there is no thickness map yet. In this case, the correction of beam hardening can be omitted first and only performed in the next iteration. However, preferably, the correction of the initial image is also performed, more specifically using a preset thickness map. For this purpose, in particular a 50% homogeneous glandular tissue portion is applied to the entire image in the breast exposure.

[0070] As mentioned above, preferably, in one embodiment of the method, a main thickness map D is formed by adding (eg three) thickness maps. The main thickness map is preferably used to correct beam hardening of the decomposed projection images and / or to form a final thickness map.

[0071] Preferably, the first decomposition is carried out with in each case N projection images and the second decomposition is carried out with projection images less than N. In particular, it is then possible to work with N×N matrices in each case.

[0072] Preferably, for the second decomposition, multiple projection images, especially two projection images, in the projection image are combined into a unified V projection image, for example, into the above-mentioned ML projection image P ML .

[0073] Preferably, for the second decomposition, a main thickness map consisting of the addition of the N thickness maps of the first decomposition is used. Alternatively or additionally, for the second decomposition, a main thickness map consisting of a model of the filmed object can be used. Alternatively or additionally, for the second decomposition, a main thickness map consisting of measurements of the filmed object can be used.

[0074] Preferably, in one embodiment of the method, energetically adjacent projection images are combined into a V projection image, but this is not mandatory. “Energically adjacent” here means that the recording energies are directly adjacent to each other. At the preferred three recording energies, the M projection images P M and L projected image P L Combined into ML projection image P ML .

[0075] The combination is preferably performed by weighted averaging of the image values ​​according to the intensity of the irradiation with the corresponding recording energy. For this purpose, the image values ​​at the same position are preferably added to each other and then normalized by means of the sum of the corresponding intensities. Such weighted averaging is essentially prior art.

[0076] Preferably, in one embodiment of the method, for the matrix for the second decomposition, the attenuation coefficients corresponding to the recording energies of the combined projection images are additionally combined, preferably by means of a weighted average. As described above, at three recording energies and in the case of combining the M projection image and the L projection image into one ML projection image, the coefficients μ M and μ L Combination is μ ML Preferably, at three recording energies, the intensity of the radiation with the corresponding recording energy is determined from the coefficient μ of the corresponding material. M and μ L The coefficient MLμ is calculated by taking the weighted average of

[0077] Preferably, in one embodiment of the method, a certain number of thickness maps and / or the main thickness map are additionally denoised. This is preferably done by means of a method that takes into account the variable thickness of the recorded object and in particular assumes a continuous thickness change. In this case, denoising preferably takes place only in the area of ​​the (recorded) object.

[0078] The device preferably comprises a correction unit which is designed to correct the projection image and / or the attenuation factor, in particular to correct beam hardening.

[0079] A preferred device comprises a denoising unit designed to denoise the thickness map. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] The present invention will be further described below based on an embodiment with reference to the drawings. In this case, the same components are provided with the same reference numerals in different drawings. The drawings are generally not to scale. The drawings show:

[0081] Figure 1 A radiographic system is shown having a device according to the invention,

[0082] Figure 2 shows a graph with attenuation and shot energy,

[0083] Figure 3 An example of a method according to the invention for estimating material thickness in radiological projection images is shown. DETAILED DESCRIPTION

[0084] Figure 1 An imaging device 1 in the form of an X-ray system 1 with a control device 2 is shown schematically. The control device 2 is equipped with a device 6 for carrying out the method according to the invention. The X-ray system 1 has a radiation source 3 in the usual manner, which is an X-ray source in this case, and during the recording of a projection image, irradiates a patient P with a beam collimated by a collimator 5, so that the radiation reaches a detector 4 opposite the radiation source 3.

[0085] Only the components that are essential for explaining the present invention are shown in the control device 2. In principle, the radiography system 1 and the associated control device 2 are known to a person skilled in the art, so that a detailed explanation is not necessary.

[0086] The control device 2 comprises a device 6 for estimating material thickness in a radiological projection image. The device 6 comprises a data interface 7, a decomposition unit 8, a correction unit 9 and a denoising unit 10.

[0087] The data interface 7 is used to receive a plurality of projection images N in number at respectively different shooting energies, preferably, an H projection image that has been shot at the highest shooting energy among three shooting energies, an M projection image that has been shot at an intermediate shooting energy among the three shooting energies, and an L projection image that has been shot at the lowest shooting energy among the three shooting energies (see also FIG. 4 ).

[0088] The decomposition unit 8 is used to perform two consecutive decompositions, namely, a first decomposition of N projection images into N thickness maps, wherein the N thickness maps respectively represent the thicknesses of N regions with respectively different materials, and a second decomposition of a certain number of main thickness maps based on the N thickness maps and / or based on corresponding measurements and a certain number of projection images among the N projection images into N result thickness maps, wherein the N result thickness maps respectively represent the thicknesses of N regions with respectively different materials.

[0089] The correction unit 9 is designed to correct the projection image. In this case, a first decomposition of the projection image into thickness maps can be performed multiple times, and the thickness maps are used to correct the beam hardening of the decomposed projection image. This can be done iteratively, that is, the correction is improved with the thickness maps currently determined, which are based on the improved decomposition of the projection image.

[0090] The denoising unit 10 is designed to denoise the thickness map. The thickness map can be a thickness map obtained by the first decomposition or a main thickness map. The denoising is preferably performed by means of a method that takes into account the variable thickness of the captured object and in particular assumes a continuous thickness change, wherein the denoising preferably takes place only in the area of ​​the object.

[0091] Figure 2 A diagram with attenuation for X-ray radiation and three recording energies EH, EM, EL is shown. For example, the diagram with a solid line shows the iodine attenuation, the diagram with a dashed line shows the glandular attenuation, and the dash-dot line diagram shows the fat attenuation. The K-edge is clearly visible at the iodine attenuation. Now, projection images are recorded at three (narrow) recording energies EH, EM, EL. At the lowest recording energy EL, the iodine attenuation and the glandular attenuation are almost identical, while the fat attenuation differs from both. At the intermediate recording energy EH, the three attenuations are clearly separated from each other. The highest recording energy EH is located in the K-edge of iodine. Therefore, there the iodine attenuation differs significantly from the other two.

[0092] If iodine is selected as material 3, then for example for the matrix used for the decomposition, the attenuation coefficient μ can be selected precisely there from the iodine graph. In the example described, it is shown that with respect to FIG. 4 (with iodine as the third material), the coefficient μ3 can be selected there H .

[0093] Figure 3 A method for estimating material thickness in a radiological projection image B is shown. The leftmost image is represented by a P vector (P L , P M , P H ). There are three projection images B, which have image values ​​at the provided coordinates (x, y). Although image B is mentioned below, it can be assumed that all calculations are performed using its image points.

[0094] P H represents the H projection image captured at the highest capture energy among the three capture energies, P M represents the M projection image captured at the intermediate capture energy, P L Indicates the L-projection image that has been captured at the lowest capture energy.

[0095] Now, by Figure 2 The attenuation coefficient μ obtained in creates the M matrix X, and the coefficient μ is used M and μ L (Weighted) combination into μ ML Create K matrices X. The two matrices X are then:

[0096] and

[0097] The two matrices X are transformed into matrices M* or K*, and the graph K (d-vector (d1, d2, d3)) is first calculated from the P-vector and the matrix M* using d=M*·P.

[0098] Now, the thickness map (d1, d2, d3) of the d-vector can be used to correct the projection image, if necessary using the main thickness map D=d1+d2+d3. This is done, for example, using the formula P korr =c0(d2 / D)+c1(d2 / D)P+c2(d2 / D)P 2 +c3(d2 / D)P 3 Then, the corrected value P korr is the new value for the P vector, from which the graph K is again derived.

[0099] Now, using the intermediate value Z, i.e. the main thickness map D, H projection image P H and as L projected image P L and M projected image P M The combined V projection image P ML , a second decomposition is performed. Now, the PD vector represented by the intermediate value Z is multiplied by the matrix K* and the final thickness maps df1, df2 and df3 are obtained as map K using df=K*·PD.

[0100] Finally, it should be pointed out again that the invention described in detail above is only an embodiment, which can be modified in different ways by a person skilled in the art without departing from the scope of the invention. In addition, the use of the indefinite article "a" or "an" does not exclude that the features involved can also exist in multiple forms. Similarly, concepts such as "unit" do not exclude that the components involved are composed of multiple co-acting sub-components, which can also be distributed in space if necessary. The term "a certain number" should be understood as "at least one". Regardless of the grammatical gender of a particular term, people with male or female gender identity are included.

Claims

1. A method for estimating material thickness in a radiological projection image, the method comprising the following steps: - providing a plurality of projection images, N in number, at respectively different capture energies, - performing a first decomposition of the N projection images into N thickness maps, the N thickness maps respectively representing the thickness of N regions of respectively different materials, - Performing a second decomposition of a certain number of main thickness maps based on the N thickness maps and / or based on corresponding measurements and a certain number of projection images among the N projection images into N resulting thickness maps, wherein the N resulting thickness maps respectively represent the thickness of N regions having respectively different materials.

2. The method according to claim 1, wherein the projection image P is decomposed into a thickness map d using a preset N×N matrix (X), preferably, wherein the matrix (X) is an attenuation coefficient (μ) μ of the material for different energies (E) ij The inverse matrix (X) of the matrix (X) formed, preferably, where - providing an H projection image, which has been acquired at the highest acquisition energy of three acquisition energies, - providing an M projection image, which has been acquired at an intermediate acquisition energy among the three acquisition energies, and - providing an L projection image, which has been captured at the lowest capture energy among the three capture energies, And the three projection images are decomposed into three thickness maps.

3. The method according to claim 1 or 2, wherein one of the thickness maps is - a K thickness map representing the thickness of the contrast agent region, and / or - a G thickness map representing the thickness of the glandular tissue, and / or - an A thickness map, the A thickness map representing the thickness of the adipose tissue, and / or - a B thickness map, said B thickness map representing bone thickness, and / or - a W thickness map, which represents the thickness of the area containing water, Preferably, the three thickness maps are preferably a K thickness map, a G thickness map and an A thickness map.

4. The method according to claim 1 , wherein the recording energy is based on the attenuation properties of the materials on which the thickness map is based. Preferably, wherein the radiation energy is selected at which the difference in the attenuation characteristics of all materials or at least one material relative to the other materials is maximal, Preferably, wherein at least one imaging energy is selected such that it is located in the descending portion of the absorption edge of one of the materials, Preferably, the lowest imaging energy is between 10 keV and 100 keV, the middle imaging energy is between 20 keV and 130 keV, and the highest imaging energy is between 30 keV and 150 keV.

5. A method according to any of the preceding claims, wherein a first decomposition of the projection image into a thickness map is performed multiple times and the thickness map is used to correct the beam hardening of the decomposed projection image and / or the coefficients of a matrix (X) used in the decomposition, wherein a polynomial correction is preferably applied.

6. A method according to any of the preceding claims, wherein a main thickness map D is formed by adding the thickness maps, and the main thickness map D is preferably used to correct beam hardening of the decomposed projection images and / or to form a final thickness map.

7. The method according to any one of the preceding claims, wherein the first decomposition is performed using respective N projection images, and the second decomposition is performed using less than N projection images, wherein preferably for the second decomposition a plurality of projection images, in particular two projection images, of the projection images are combined into a unified V projection image, preferably, wherein for the second decomposition a main thickness map constructed as follows is used, the main thickness map - is formed by the addition of N thickness maps of said first decomposition, and / or - consist of a model of the object being photographed, and / or - consists of measurements of the photographed object.

8. The method according to claim 7 , wherein energetically adjacent projection images are combined to form a V projection image, in particular at three recording energies, an M projection image and an L projection image are combined to form an ML projection image, preferably by means of a weighted averaging of the image values ​​according to the intensity (I) of the irradiation with the corresponding recording energy, Preferably, the image values ​​at the same position are added to one another and then normalized by means of the sum of the corresponding intensities (I), or the image values ​​normalized with the corresponding intensities (I) are added.

9. A method according to claim 8, wherein for the matrix (X) used for the second decomposition, those attenuation coefficients (μ) corresponding to the shooting energies of the combined projection images are additionally combined, preferably by means of weighted averaging, preferably wherein at three shooting energies the coefficient μML is calculated from the weighted average of the coefficients μM and μL for the corresponding materials according to the intensity (I) of the irradiation with the corresponding shooting energies.

10. A method according to any of the preceding claims, wherein a certain number of thickness maps and / or main thickness maps are additionally denoised, preferably with the aid of a method which takes into account the variable thickness of the photographed object and in particular assumes a continuous thickness change, wherein the denoising preferably takes place only in the region of the object.

11. A device (6) for estimating material thickness in a radiological projection image, comprising: - a data interface (7) for receiving a plurality of projection images, N in number, at respectively different recording energies, preferably an H projection image captured at the highest recording energy among three recording energies, an M projection image captured at an intermediate recording energy among the three recording energies, and an L projection image captured at the lowest recording energy among the three recording energies, - a decomposition unit (8), the decomposition unit being designed to: - a first decomposition of the N projection images into N thickness maps, the N thickness maps respectively representing the thickness of N regions of respectively different materials, and - a second decomposition based on the N thickness maps and / or on a number of main thickness maps based on corresponding measurements and a number of the N projection images into N resulting thickness maps, the N resulting thickness maps respectively representing the thickness of N regions of respectively different materials, Preferably, the three projection images are decomposed into three thickness maps, wherein the three thickness maps respectively represent the thicknesses of three different materials.

12. The device (6) according to claim 11, comprising - a correction unit (9) designed to correct the projection image and / or the attenuation coefficient (μ), and / or A denoising unit (10) is designed to denoise the thickness map.

13. An imaging device (1), preferably a multi-slice detector or a photon counting detector, designed for spectrographic recording of an object by means of radiation, comprising: A radiation source (3) designed to emit a spectrum comprising N different imaging energies (EH, EM, EL); A detector unit (4) designed to detect the N acquisition energies (EH, EM, EL); and a device (6) according to any one of claims 11 or 12.

14. A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to perform the steps of the method according to claims 1 to 10.

15. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method according to claims 1 to 10.

Images