CN117252944A - Decomposition method and system for dual-energy spectrum CT reconstructed image - Google Patents
Decomposition method and system for dual-energy spectrum CT reconstructed image Download PDFInfo
- Publication number
- CN117252944A CN117252944A CN202311211997.0A CN202311211997A CN117252944A CN 117252944 A CN117252944 A CN 117252944A CN 202311211997 A CN202311211997 A CN 202311211997A CN 117252944 A CN117252944 A CN 117252944A
- Authority
- CN
- China
- Prior art keywords
- image
- projection data
- energy spectrum
- dual
- decomposition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001228 spectrum Methods 0.000 title claims abstract description 147
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 130
- 238000000034 method Methods 0.000 title claims abstract description 59
- 239000000463 material Substances 0.000 claims abstract description 268
- 230000004927 fusion Effects 0.000 claims abstract description 20
- 230000006870 function Effects 0.000 claims description 19
- 238000011156 evaluation Methods 0.000 claims description 18
- 230000005855 radiation Effects 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 8
- 230000009977 dual effect Effects 0.000 claims description 7
- 230000003595 spectral effect Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 19
- 239000000126 substance Substances 0.000 abstract description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000005461 Bremsstrahlung Effects 0.000 description 2
- 210000000988 bone and bone Anatomy 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- NNJPGOLRFBJNIW-HNNXBMFYSA-N (-)-demecolcine Chemical compound C1=C(OC)C(=O)C=C2[C@@H](NC)CCC3=CC(OC)=C(OC)C(OC)=C3C2=C1 NNJPGOLRFBJNIW-HNNXBMFYSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000013170 computed tomography imaging Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- SYHGEUNFJIGTRX-UHFFFAOYSA-N methylenedioxypyrovalerone Chemical compound C=1C=C2OCOC2=CC=1C(=O)C(CCC)N1CCCC1 SYHGEUNFJIGTRX-UHFFFAOYSA-N 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 210000004872 soft tissue Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2210/00—Indexing scheme for image generation or computer graphics
- G06T2210/41—Medical
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
The invention discloses a decomposition method and a system of a double-energy spectrum CT reconstructed image, comprising the following steps: acquiring dual-energy spectrum projection data and presetting material projection data to be solved; constructing a polynomial according to the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved; constructing a material template image according to the dual-energy spectrum projection data; constructing a material decomposition image containing a preset algebra according to the dual-energy spectrum projection data and the polynomial; solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients; solving the polynomial according to the target coefficients to obtain decomposed projection data of a plurality of materials; reconstructing an image according to the decomposed multiple material projection data to obtain a multi-material fusion image. The decomposition process of the double-energy spectrum CT reconstruction image on different substances is simplified, the universality of the decomposition method on different application scenes is improved, and the calculation cost is reduced.
Description
Technical Field
The invention relates to the technical field of X-ray multi-energy spectrum CT imaging, in particular to a decomposition method and a decomposition system of a double-energy spectrum CT reconstructed image.
Background
X-rays are generated by characteristic radiation and bremsstrahlung, wherein bremsstrahlung generates a continuous energy spectrum, and therefore X-rays are not single energies. Because the attenuation coefficients of substances under different energies are different, hardening artifacts and cup-shaped artifacts can be generated in the reconstructed X-ray image, and the diagnosis of doctors is interfered; on the other hand, different materials can be distinguished by using the attenuation characteristics of different substances at different energies, which is the physical basis of dual-energy spectrum image reconstruction.
The existing dual-energy spectrum decomposition technology can be divided into three main categories, namely decomposition based on a projection domain, decomposition based on an image domain and iterative decomposition. The method is characterized in that the high-energy spectrum information and the low-energy spectrum information in the X-rays can be fully utilized based on the decomposition of a projection domain, the nonlinear problem is converted into the linear problem to solve, and different materials are separated in the projection domain, so that images of different materials can be respectively reconstructed by directly using a reconstruction algorithm of a conventional image, hardening artifacts can be effectively eliminated, the calculation is simple and convenient, the efficiency is high, and the method has high requirement on the spatial consistency of projection data. The image domain decomposition is based on the fact that on the basis of reconstructing an image by using high-energy spectrum information and low-energy spectrum information, different materials are solved by using attenuation coefficient differences of the different materials, and the requirement on spatial consistency of projection data is not high. The iterative decomposition is to carry out statistical accumulation on attenuation information under different energy levels, so that mathematical description of an imaging model is nonlinear, solving is more complex, calculation speed is low, and practicability is poor.
In addition, conventional dual-energy CT systems are high and low energy single sampling, requiring the patient to receive two doses of radiation, and it is difficult to ensure spatial consistency of the projection data.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the decomposition method and the system for the double-energy-spectrum CT reconstructed image are provided, the decomposition process of the double-energy-spectrum CT reconstructed image on different substances is simplified, the universality of the decomposition method on different application scenes is improved, and the calculation cost is reduced.
In order to solve the technical problems, the invention adopts the following technical scheme:
a decomposition method of a dual energy spectrum CT reconstructed image, comprising:
acquiring dual-energy spectrum projection data and presetting material projection data to be solved;
constructing a polynomial according to the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved;
constructing a material template image according to the dual-energy spectrum projection data;
constructing a material decomposition image containing a preset algebra according to the dual-energy spectrum projection data and the polynomial;
solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients;
solving the polynomial according to the target coefficients to obtain decomposed projection data of a plurality of materials;
reconstructing an image according to the decomposed multiple material projection data to obtain a multi-material fusion image.
In order to solve the technical problems, the invention adopts another technical scheme that:
a decomposition system of a dual-energy spectrum CT reconstructed image comprises a ray source generator, a top layer detector, a bottom layer detector, a rotary table and an image decomposition terminal; the radiation source generator is arranged on one side of the top layer detector, and the emitting end of the radiation source generator is arranged opposite to the top layer detector; the bottom layer detector is arranged on one side of the top layer detector away from the ray source generator; the rotary table is arranged between the ray source generator and the top-layer detector; the image decomposition terminal is respectively and electrically connected with the top layer detector and the bottom layer detector;
the image decomposition terminal comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the following steps:
acquiring dual-energy spectrum projection data and presetting material projection data to be solved;
constructing a polynomial according to the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved;
constructing a material template image according to the dual-energy spectrum projection data;
constructing a material decomposition image containing a preset algebra according to the dual-energy spectrum projection data and the polynomial;
solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients;
solving the polynomial according to the target coefficients to obtain decomposed projection data of a plurality of materials;
reconstructing an image according to the decomposed multiple material projection data to obtain a multi-material fusion image.
The invention has the beneficial effects that: and constructing an energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved into a polynomial, constructing a material template image by acquiring the dual-energy spectrum projection data, constructing a material decomposition image containing a preset algebra according to the polynomial, solving unknown values of the coefficients in the material decomposition image according to the known material template image, finally solving the polynomial by taking the values of the coefficients as unknown target coefficients in the polynomial, obtaining a plurality of decomposed material projection data, and finally carrying out image reconstruction according to the plurality of material projection data, thereby obtaining the multi-material fusion image with different densities. According to the method, a polynomial is constructed to avoid calculation errors caused by forced solution of unknown physical quantities when material projection data are decomposed based on an energy attenuation relation, and the original complex physical quantity solving process is converted into polynomial target coefficient solving, so that the complexity of a dual-energy spectrum decomposing process is effectively reduced, and the calculation cost is greatly saved; and the energy spectrum information is not needed to be used as priori information, and the image is not needed to be subjected to water pre-correction, so that the universality of the decomposition method on different application scenes is improved.
Drawings
FIG. 1 is a flow chart of steps of a method for decomposing a dual-energy spectrum CT reconstructed image according to an embodiment of the present invention;
FIG. 2 is a program flow chart of a method for decomposing a dual-energy spectrum CT reconstructed image according to an embodiment of the present invention;
FIG. 3 is an effect diagram of an exploded method of a dual-energy spectrum CT reconstructed image according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a decomposition system for reconstructing an image by dual-energy spectrum CT according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an image decomposition terminal according to an embodiment of the present invention;
description of the reference numerals:
100. a radiation source generator; 200. a top layer detector; 300. a bottom layer detector; 400. a rotary table; 500. an image decomposition terminal; 501. a memory; 502. a processor.
Detailed Description
In order to describe the technical contents, the achieved objects and effects of the present invention in detail, the following description will be made with reference to the embodiments in conjunction with the accompanying drawings.
Referring to fig. 1, an embodiment of the present invention provides a decomposition method of a dual-energy spectrum CT reconstructed image, including:
acquiring dual-energy spectrum projection data and presetting material projection data to be solved;
constructing a polynomial according to the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved;
constructing a material template image according to the dual-energy spectrum projection data;
constructing a material decomposition image containing a preset algebra according to the dual-energy spectrum projection data and the polynomial;
solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients;
solving the polynomial according to the target coefficients to obtain decomposed projection data of a plurality of materials;
reconstructing an image according to the decomposed multiple material projection data to obtain a multi-material fusion image.
From the above description, the beneficial effects of the invention are as follows: and constructing an energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved into a polynomial, constructing a material template image by acquiring the dual-energy spectrum projection data, constructing a material decomposition image containing a preset algebra according to the polynomial, solving unknown values of the coefficients in the material decomposition image according to the known material template image, finally solving the polynomial by taking the values of the coefficients as unknown target coefficients in the polynomial, obtaining a plurality of decomposed material projection data, and finally carrying out image reconstruction according to the plurality of material projection data, thereby obtaining the multi-material fusion image with different densities. According to the method, a polynomial is constructed to avoid calculation errors caused by forced solution of unknown physical quantities when material projection data are decomposed based on an energy attenuation relation, and the original complex physical quantity solving process is converted into polynomial target coefficient solving, so that the complexity of a dual-energy spectrum decomposing process is effectively reduced, and the calculation cost is greatly saved; and the energy spectrum information is not needed to be used as priori information, and the image is not needed to be subjected to water pre-correction, so that the universality of the decomposition method on different application scenes is improved.
Further, the dual-energy spectrum projection data comprises high-energy spectrum projection data and low-energy spectrum projection data;
the material template image comprises a high-density material template image and a low-density material template image;
the constructing a material template image from the dual-energy spectrum projection data comprises:
selecting the high-energy spectrum projection data or the low-energy spectrum projection data for image reconstruction to obtain a base image;
and dividing the base image to obtain a high-density material template image and a low-density material template image.
As can be seen from the above description, any energy spectrum projection data is selected to perform image reconstruction to obtain a base image, and a template image is segmented on the basis of the base image, so that a known data amount is provided for a preset value of a subsequent solving material decomposition image, and the calculated value of the coefficient is conveniently used as a target coefficient to solve a polynomial; and meanwhile, a template image is acquired based on the base image so as to ensure the accuracy of solving coefficients.
Further, the constructing a material decomposition image including a preset algebra according to the dual-energy spectrum projection data and the polynomial includes:
performing power operation according to the dual-energy spectrum projection data to obtain a plurality of power projection data;
respectively carrying out back projection reconstruction on the multiple power projection data to obtain an intermediate image;
and constructing a material decomposition image containing the preset algebra according to the preset algebra and the intermediate image.
As can be seen from the above description, the multiple power projection data are constructed by performing power operation on the dual-energy spectrum projection data, so that the dual-energy spectrum projection data are converted into corresponding polynomial data in the polynomial, and meanwhile, an intermediate image is obtained after the back projection reconstruction of the power projection data, and a material decomposition image is constructed by the intermediate image and a preset algebra. At this time, since the known material template image and the material decomposition image need to keep an equal relation, an equation for solving a preset algebra is established, and since the finally obtained data type is an image, the accuracy of the construction of the final image is ensured by carrying out the solving operation in an image mode.
Further, the solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients includes:
constructing an evaluation function according to the material template image and the material decomposition image;
and solving the value of the preset algebra according to the evaluation function to obtain a plurality of target coefficients.
As can be seen from the above description, the material projection data belongs to an unknown data amount in the polynomial, and the target coefficient used for representing the energy attenuation relation between the material projection data and the dual-energy spectrum projection data in the polynomial is also in an unknown state, that is, the material projection data cannot be solved if the polynomial contains two unknown amounts in the projection domain. The material decomposition image can be obtained after the projection data are subjected to back projection reconstruction, so that the material decomposition image is converted into the image relationship respectively after the back projection reconstruction is performed on the basis of the projection data relationship represented in the polynomial, a template image can be obtained in an image domain and used as the known quantity of the material projection data in the polynomial, a preset algebra is used as a target coefficient in the polynomial, an intermediate image is used as dual-energy spectrum projection data in the polynomial, the preset algebra in the material decomposition image is solved by constructing an evaluation function, and the target coefficient of the polynomial can be obtained, so that the material projection data is solved by the polynomial. The unknown quantity is converted into the known quantity by means of back projection reconstruction image, so that unknown target coefficients in the polynomial are solved, and a complex calculation process is avoided.
Further, reconstructing an image according to the decomposed plurality of material projection data to obtain a multi-material fusion image includes:
respectively carrying out back projection reconstruction according to the decomposed multiple material projection data to obtain multiple target decomposed images;
and linearly combining the target decomposition images to obtain a multi-material fusion image.
As can be seen from the above description, after the decomposed multiple material projection data are obtained by polynomial solving, multiple target decomposed images can be obtained by respectively performing back projection reconstruction on the multiple material projection data, and a multiple material fusion image can be obtained by linearly combining the target decomposed images, so that hardening artifacts existing in the original image are effectively corrected.
Further, the constructing an evaluation function according to the material template image and the material decomposition image includes:
constructing an evaluation function according to the material template image and the material decomposition image by a least square method:
E 2 ＝∫(f(r)-t(r)) 2 d 2 r；
wherein f (r) represents a material decomposition image, and t (r) represents a material template image; r represents the line integral of the path corresponding to s+λθ, s represents the source vector, θ is the projection angle, and λ is the line integral along θ; e represents an error value.
From the above description, it can be seen that the evaluation function is constructed by the least square method, that is, the optimal solution of the preset algebra can be obtained by solving the minimum error between the material decomposition image and the material template image. In this way, the target coefficient of the polynomial can be simply solved, and the increase of the complexity of calculation is avoided.
Referring to fig. 4, another embodiment of the present invention provides a dual-energy spectrum CT image reconstruction system, which includes a radiation source generator, a top detector, a bottom detector, a rotation table, and an image decomposition terminal; the radiation source generator is arranged on one side of the top layer detector, and the emitting end of the radiation source generator is arranged opposite to the top layer detector; the bottom layer detector is arranged on one side of the top layer detector away from the ray source generator; the rotary table is arranged between the ray source generator and the top-layer detector; the image decomposition terminal is respectively and electrically connected with the top layer detector and the bottom layer detector;
the image decomposition terminal comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the following steps:
acquiring dual-energy spectrum projection data and presetting material projection data to be solved;
constructing a polynomial according to the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved;
constructing a material template image according to the dual-energy spectrum projection data;
constructing a material decomposition image containing a preset algebra according to the dual-energy spectrum projection data and the polynomial;
solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients;
solving the polynomial according to the target coefficients to obtain decomposed projection data of a plurality of materials;
reconstructing an image according to the decomposed multiple material projection data to obtain a multi-material fusion image.
From the above description, the beneficial effects of the invention are as follows: the radiation source generator, the top layer detector and the bottom layer detector are combined, the top layer detector obtains low-energy spectrum projection data, the bottom layer detector obtains high-energy rays after secondary attenuation, namely high-energy spectrum projection data, and the high-energy spectrum projection data and the low-energy spectrum projection data can be obtained simultaneously through one-time scanning, so that the spatial consistency of the projection data is ensured. In addition, a polynomial is constructed to avoid calculation errors caused by forced solution of unknown physical quantities when material projection data are decomposed based on an energy attenuation relation, and the original complex physical quantity solving process is converted into a polynomial target coefficient solving process, so that the complexity of a dual-energy spectrum decomposing process is effectively reduced, and the calculation cost is greatly saved; and the energy spectrum information is not needed to be used as priori information, and the image is not needed to be subjected to water pre-correction, so that the universality of the decomposition method on different application scenes is improved.
Further, the dual-energy spectrum projection data comprises high-energy spectrum projection data and low-energy spectrum projection data;
the material template image comprises a high-density material template image and a low-density material template image;
the constructing a material template image from the dual-energy spectrum projection data comprises:
selecting the high-energy spectrum projection data or the low-energy spectrum projection data for image reconstruction to obtain a base image;
and dividing the base image to obtain a high-density material template image and a low-density material template image.
As can be seen from the above description, any energy spectrum projection data is selected to perform image reconstruction to obtain a base image, and a template image is segmented on the basis of the base image, so that a known data amount is provided for a preset value of a subsequent solving material decomposition image, and the calculated value of the coefficient is conveniently used as a target coefficient to solve a polynomial; and meanwhile, a template image is acquired based on the base image so as to ensure the accuracy of solving coefficients.
Further, the constructing a material decomposition image including a preset algebra according to the dual-energy spectrum projection data and the polynomial includes:
performing power operation according to the dual-energy spectrum projection data to obtain a plurality of power projection data;
respectively carrying out back projection reconstruction on the multiple power projection data to obtain an intermediate image;
and constructing a material decomposition image containing the preset algebra according to the preset algebra and the intermediate image.
As can be seen from the above description, the multiple power projection data are constructed by performing power operation on the dual-energy spectrum projection data, so that the dual-energy spectrum projection data are converted into corresponding polynomial data in the polynomial, and meanwhile, an intermediate image is obtained after the back projection reconstruction of the power projection data, and a material decomposition image is constructed by the intermediate image and a preset algebra. At this time, since the known material template image and the material decomposition image need to keep an equal relation, an equation for solving a preset algebra is established, and since the finally obtained data type is an image, the accuracy of the construction of the final image is ensured by carrying out the solving operation in an image mode.
Further, the solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients includes:
constructing an evaluation function according to the material template image and the material decomposition image;
and solving the value of the preset algebra according to the evaluation function to obtain a plurality of target coefficients.
As can be seen from the above description, the material projection data belongs to an unknown data amount in the polynomial, and the target coefficient used for representing the energy attenuation relation between the material projection data and the dual-energy spectrum projection data in the polynomial is also in an unknown state, that is, the material projection data cannot be solved if the polynomial contains two unknown amounts in the projection domain. The material decomposition image can be obtained after the projection data are subjected to back projection reconstruction, so that the material decomposition image is converted into the image relationship respectively after the back projection reconstruction is performed on the basis of the projection data relationship represented in the polynomial, a template image can be obtained in an image domain and used as the known quantity of the material projection data in the polynomial, a preset algebra is used as a target coefficient in the polynomial, an intermediate image is used as dual-energy spectrum projection data in the polynomial, the preset algebra in the material decomposition image is solved by constructing an evaluation function, and the target coefficient of the polynomial can be obtained, so that the material projection data is solved by the polynomial. The unknown quantity is converted into the known quantity by means of back projection reconstruction image, so that unknown target coefficients in the polynomial are solved, and a complex calculation process is avoided.
The embodiment of the invention provides a decomposition method and a decomposition system of a dual-energy spectrum CT reconstructed image, which can be applied to a medical X-ray image processing scene, simplify the decomposition process of the dual-energy spectrum CT reconstructed image on different substances, improve the universality of the decomposition method on different application scenes and reduce the calculation cost, and the following is explained by a specific embodiment:
referring to fig. 1 to 3, a first embodiment of the present invention is as follows:
a decomposition method of a dual energy spectrum CT reconstructed image, comprising:
s1, acquiring dual-energy spectrum projection data and presetting material projection data to be solved.
Wherein the dual-energy spectrum projection data includes high-energy spectrum projection data and low-energy spectrum projection data.
S2, constructing a polynomial according to the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved.
In some embodiments, the polynomial constructed according to the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved is specifically:
wherein D is i (q 1 ,q 2 ) Representation comprising dual energy spectrum projection data q 1 And q 2 Polynomial of p i Representing projection data of material, b in (q 1 ,q 2 ) The power projection data corresponding to the dual-energy spectrum projection data is represented, and i represents the numbers of different kinds of materials;q 1 and q 2 Respectively representing high-energy spectrum projection data and low-energy spectrum projection data, k and l respectively representing q 1 And q 2 Where k=0, 1,2, …, K, l=0, 1,2, …, L, and n=k (l+1) +l, N represents the total number of polynomial numbers, and n= (k+1) (l+1); c in The target coefficient to be solved is also a preset algebra.
In some embodiments, the energy attenuation relation between the dual-energy spectrum projection data and the projection data of the material to be solved is specifically:
wherein q j Representing dual energy spectrum projection data, omega j (E) Representing normalized energy, p, in relation to energy 1 And p 2 Material projection data, ψ, representing two different materials, respectively 1 (E) And psi is equal to 2 (E) Two energy-related parameters are represented, respectively, and E represents energy.
The energy attenuation relationship between the dual-energy spectrum projection data q and the material projection data p, i.e., q, can be known by the formula (3) j ＝q j (p 1 ,p 2 ) In this case, the conventional decomposition method is to solve the material projection data p in the projection domain by the dual-energy spectrum projection data q in the opposite direction, i.e. p i ＝p i (q 1 ,q 2 ) And then obtaining a target decomposition image through back projection reconstruction. But ω in equation (3) j (E)、Ψ 1 (E) And psi is equal to 2 (E) Is difficult to measure and quantify, and in order to avoid erroneous data caused by forcibly solving the parameters, the invention adopts a polynomial (namely, formula (1)) containing q and p to replace formula (3) for solving.
In some embodiments, the specific method for obtaining the energy attenuation relation between the dual-energy spectrum projection data and the material projection data is as follows:
since X-rays are not single energy, but multi-energy, their multi-energy spectral attenuation formula can be expressed as follows:
where L is the line integral of the path corresponding to s+λθ, s is the source vector, θ is the projection angle, λ is the line integral along direction θ, E represents the energy, ω (L, E) is the normalized energy associated with the path, energy, and the energy attenuation on path L can be represented by μ (E, s+λθ). Subsequently denoted by r, μ (E, s+λθ) can also be written as μ (E, r), which is the attenuation of the multi-spectral radiation after it has passed through the object, which attenuation is a function of both the object and the energy, and therefore also the substance-related parameter f 0 (s+λθ) and an energy-related parameter ψ 0 (E) Two physical quantities represent the attenuation values, namely:
μ(E,s+λθ)＝f 0 (s+λθ)ψ 0 (E)
＝f 0 (r)ψ 0 (E) (5)；
wherein, psi is 0 (E) Is an energy dependent term of the material in the object, equation (4) can be written as:
order theThe above equation (6) can be written as:
the above formula (7) corresponds to the replacement of one parameter mu related to the energy and the radiation passing distance by a parameter ψ related to the energy respectively 0 (E) And a parameter p0 related to the ray passing distance. Assuming that the object contains two materials (water and bone, which differ significantly in density), then substance f 0 (r) and energy ψ 0 (E) A kind of electronic deviceThere will be two dependent terms, and equation (5) above can be written as:
the final solution to be found is the material image f 1 (r) and f 2 (r), i.e., the decomposed material image, according to equation (8), equation (7) can be written as:
and obtaining the energy attenuation relation between the dual-energy spectrum projection data and the projection data of the material to be solved.
S3, constructing a material template image according to the dual-energy spectrum projection data.
Wherein the material template image comprises a high density material template image and a low density material template image.
Specifically, the step S3 includes:
s31, selecting the high-energy spectrum projection data or the low-energy spectrum projection data for image reconstruction to obtain a base image;
in some embodiments, the image reconstruction may employ a CT image reconstruction method such as filtered back projection reconstruction (FPB), back projection filtered reconstruction (BPF), iterative reconstruction, and the like.
S32, dividing the base image to obtain a high-density material template image and a low-density material template image.
In some embodiments, the base image includes a material containing a high density material (e.g., bone) and a material containing a low density material (e.g., soft tissue), so the material template image includes a high density material template image and a low density material template image.
In an alternative embodiment, the template image may be segmented according to the gray values of the base image, and the image may be segmented according to the difference in gray values due to the different gray values displayed by the different density materials in the reconstructed base image. In some embodiments, specifically: presetting a high gray threshold and a low gray threshold, and dividing an image with a gray value larger than the high gray threshold in the base image into high-density material template images; and dividing an image with the gray value larger than the low gray threshold value and smaller than the high gray threshold value in the base image into low-density material template images. It should be noted that, in order to facilitate subsequent calculation, the pixel of the image to be segmented in the base image may be set to 1, and the pixels of other images not to be segmented may be set to 0, so as to achieve the image segmentation effect.
In an alternative embodiment, the template image may be segmented according to the gray level variation of the image edge, and since the different materials have edges with strong variation in the adjacent areas, the gradient difference may be used to find the image edge to segment the different materials.
In an alternative embodiment, the template image may be obtained using a region-based segmentation method, and regions of similar nature may be found and the same material may be segmented.
S4, constructing a material decomposition image containing a preset algebra according to the dual-energy spectrum projection data and the polynomial.
Specifically, the step S4 includes:
s41, performing power operation according to the dual-energy spectrum projection data to obtain a plurality of power projection data;
in some embodiments, the step S41 is specifically: and (3) performing power operation on the dual-energy spectrum projection data according to the formula (2) to obtain a plurality of power projection data.
S42, respectively carrying out back projection reconstruction on the plurality of power projection data to obtain an intermediate image;
s43, constructing a material decomposition image containing the preset algebra according to the preset algebra and the intermediate image.
In some embodiments, the material projection data is back-projection reconstructed to obtain a material decomposition image, and thus, according to the above formula (1):
wherein,representing back projection, f (r) representing a material decomposition image, f n (r) represents an intermediate image, c n Representing a preset algebra. That is, the material projection data p is subjected to back projection reconstruction to obtain a material decomposition image f (r), and as can be known from the formula (1), p=d (q 1 ,q 2 ) Then D (q 1 ,q 2 ) A material decomposition image f (r) can be obtained through back projection reconstruction, andwill->A material decomposition image f (r) can be obtained through back projection reconstruction, wherein c n Is a preset algebra and belongs to a constant; b n (q 1 ,q 2 ) Is power projection data; therefore, the corresponding material decomposition image f (r) can be obtained by only carrying out back projection reconstruction on the power projection data and multiplying the power projection data by a constant, and the power projection data b n (q 1 ,q 2 ) And obtaining an intermediate image through back projection reconstruction. The above c n Expressed as a preset algebra when unknown values, c n Expressed as a target coefficient when the value is known.
S5, solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients.
Specifically, the step S5 includes:
s51, constructing an evaluation function according to the material template image and the material decomposition image;
specifically, the step S51 includes:
s511, constructing an evaluation function according to the material template image and the material decomposition image by a least square method:
E 2 ＝∫(f(r)-t(r)) 2 d 2 r(11)；
wherein f (r) represents a material decomposition image, and t (r) represents a material template image; r represents the line integral of the path corresponding to s+λθ, s represents the source vector, θ is the projection angle, and λ is the line integral along θ; e represents an error value.
S52, solving the value of the preset algebra according to the evaluation function to obtain a plurality of target coefficients.
In some embodiments, the high density material is template image t 1 (r) and Low Density Material template image t 2 And (r) substituting the above formula (11) to obtain N target coefficients corresponding to the high-density material template image and N target coefficients corresponding to the low-density material template image, namely obtaining N target coefficients of two different-density materials in the base image, wherein the total is 2N target coefficients.
And S6, solving the polynomial according to the target coefficients to obtain decomposed projection data of a plurality of materials.
In some embodiments, N target coefficients c for two different densities of material n The polynomial can be solved by substituting the formula (1) respectively, so as to obtain the projection data p of the two decomposed materials 1 And p 2 。
S7, reconstructing an image according to the decomposed multiple material projection data to obtain a multiple material fusion image.
Specifically, the step S7 includes:
s71, respectively carrying out back projection reconstruction according to the decomposed projection data of the materials to obtain a plurality of target decomposed images;
in some embodiments, the decomposed two materials are projected into data p 1 And p 2 And respectively carrying out back projection reconstruction to obtain two target decomposition images containing materials with different densities.
S72, linearly combining the target decomposition images to obtain a multi-material fusion image.
It should be noted that, since the base image is reconstructed by selecting one kind of energy spectrum projection data, the one multi-material fusion image is a single energy spectrum reconstructed image.
In this embodiment, the low-energy spectrum projection data is selected for image reconstruction to obtain a corresponding base image, and the corresponding high-density target decomposition image and low-density target decomposition image are obtained after back projection reconstruction according to the decomposed material projection data, and a multi-material fusion image is obtained, as shown in fig. 3. As can be seen from fig. 3, the decomposition method of the present invention can decompose two materials with different densities, and the fusion of the reference image and the multi-material image can effectively correct the hardening artifact in the image.
Referring to fig. 4 to 5, a second embodiment of the present invention is as follows:
a decomposition system of a dual-energy spectrum CT reconstructed image comprises a radiation source generator 100, a top layer detector 200, a bottom layer detector 300, a rotary table 400 and an image decomposition terminal 500; the radiation source generator 100 is disposed at one side of the top detector 200, and the emitting end of the radiation source generator 100 is disposed opposite to the top detector 200; the bottom layer detector 300 is disposed on a side of the top layer detector 200 remote from the source generator 100; the rotary table 400 is disposed between the radiation source generator 100 and the top detector 200; the image decomposition terminal 500 is electrically connected to the top layer detector 200 and the bottom layer detector 300, respectively; the image decomposition terminal 500 includes a memory 501, a processor 502 and a computer program stored in the memory 501 and executable on the processor 502, wherein the processor 502 implements the steps of the decomposition method for reconstructing an image by dual energy spectrum CT of embodiment one when the computer program is executed.
As shown in fig. 4, the high-energy spectrum projection data and the low-energy spectrum projection data can be obtained simultaneously by combining the top-layer detector 200 and the bottom-layer detector 300, so that the method is simpler in structure than the traditional measurement method of a single-energy spectrum detector, and different acquisition schemes can be designed according to shooting requirements.
In some embodiments, during the scanning process, the relative positions of the radiation source generator 100, the top detector 200 and the bottom detector 300 are kept fixed, the object to be detected is placed on the rotary table 400, the rotary table 400 moves around the rotation center 401 (i.e. the rotary table rotates), and the dual-energy spectrum projection data can be obtained after scanning the specified angle range of the object to be detected. Wherein the scan angle should be not less than 180 ° fan angle. To ensure consistency of dual-energy spectral projection data, the source generator 100 employs a pulsed perspective approach.
In summary, according to the decomposition method and system for the dual-energy spectrum CT reconstructed image provided by the invention, the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved is constructed as a polynomial, and the material projection data can be directly obtained by solving the dual-energy spectrum projection data based on the polynomial, but the preset algebra exists in the polynomial, so that the material projection data can be solved only by solving the value of the preset algebra in the polynomial. On the basis, a material template image is constructed by acquiring dual-energy spectrum projection data, a material decomposition image containing preset algebra is constructed according to a polynomial, so that the value of the preset algebra in the material decomposition image is solved according to the known material template image, the projection domain calculation is converted into the image domain calculation in the mode, the value of the preset algebra is solved according to the known image in the image domain, and then the value is substituted into the polynomial of the projection domain again to solve the decomposed material projection data. And finally, reconstructing an image according to the projection data of the multiple materials to obtain the multi-material fusion image with different densities. According to the method, a polynomial is constructed to avoid calculation errors caused by forced measurement and solution of unknown physical quantities when material projection data are decomposed based on an energy attenuation relation, and an original complex physical quantity solving process is converted into a polynomial target coefficient solving process, so that the complexity of a dual-energy spectrum decomposing process is effectively reduced, and the calculation cost is greatly saved; in addition, objects with similar structures and respectively close densities of two materials can be solved by using a set of target coefficients, so that the decomposition method can be applied to two-dimensional images and three-dimensional images, and the universality of the decomposition method for different application scenes is improved. Meanwhile, the decomposition system can acquire the dual-energy spectrum projection data at the same time only by scanning once, thereby ensuring the spatial consistency of the projection data.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent changes made by the specification and drawings of the present invention, or direct or indirect application in the relevant art, are included in the scope of the present invention.
Claims (10)
1. A decomposition method of a dual-energy spectrum CT reconstructed image, comprising:
acquiring dual-energy spectrum projection data and presetting material projection data to be solved;
constructing a polynomial according to the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved;
constructing a material template image according to the dual-energy spectrum projection data;
constructing a material decomposition image containing a preset algebra according to the dual-energy spectrum projection data and the polynomial;
solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients;
solving the polynomial according to the target coefficients to obtain decomposed projection data of a plurality of materials;
reconstructing an image according to the decomposed multiple material projection data to obtain a multi-material fusion image.
2. The method of claim 1, wherein the dual-energy spectrum projection data comprises high-energy spectrum projection data and low-energy spectrum projection data;
the material template image comprises a high-density material template image and a low-density material template image;
the constructing a material template image from the dual-energy spectrum projection data comprises:
selecting the high-energy spectrum projection data or the low-energy spectrum projection data for image reconstruction to obtain a base image;
and dividing the base image to obtain a high-density material template image and a low-density material template image.
3. The method of claim 1, wherein constructing a material decomposition image comprising a predetermined algebra from the dual-spectral projection data and the polynomial comprises:
performing power operation according to the dual-energy spectrum projection data to obtain a plurality of power projection data;
respectively carrying out back projection reconstruction on the multiple power projection data to obtain an intermediate image;
and constructing a material decomposition image containing the preset algebra according to the preset algebra and the intermediate image.
4. The method of claim 1, wherein solving the values of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients comprises:
constructing an evaluation function according to the material template image and the material decomposition image;
and solving the value of the preset algebra according to the evaluation function to obtain a plurality of target coefficients.
5. The method of claim 1, wherein reconstructing an image from the decomposed plurality of material projection data to obtain a multi-material fusion image comprises:
respectively carrying out back projection reconstruction according to the decomposed multiple material projection data to obtain multiple target decomposed images;
and linearly combining the target decomposition images to obtain a multi-material fusion image.
6. The method of claim 4, wherein constructing an evaluation function from the material template image and the material decomposition image comprises:
constructing an evaluation function according to the material template image and the material decomposition image by a least square method:
E 2 ＝∫(f(r)-t(r)) 2 d 2 r；
wherein f (r) represents a material decomposition image, and t (r) represents a material template image; r represents the line integral of the path corresponding to s+λθ, s represents the source vector, θ is the projection angle, and λ is the line integral along θ; e represents an error value.
7. The decomposition system of the dual-energy spectrum CT reconstructed image is characterized by comprising a ray source generator, a top layer detector, a bottom layer detector, a rotary table and an image decomposition terminal; the radiation source generator is arranged on one side of the top layer detector, and the emitting end of the radiation source generator is arranged opposite to the top layer detector; the bottom layer detector is arranged on one side of the top layer detector away from the ray source generator; the rotary table is arranged between the ray source generator and the top-layer detector; the image decomposition terminal is respectively and electrically connected with the top layer detector and the bottom layer detector;
the image decomposition terminal comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the following steps:
acquiring dual-energy spectrum projection data and presetting material projection data to be solved;
constructing a polynomial according to the energy attenuation relation between the dual-energy spectrum projection data and the material projection data to be solved;
constructing a material template image according to the dual-energy spectrum projection data;
constructing a material decomposition image containing a preset algebra according to the dual-energy spectrum projection data and the polynomial;
solving the value of the preset algebra according to the material template image and the material decomposition image to obtain a plurality of target coefficients;
solving the polynomial according to the target coefficients to obtain decomposed projection data of a plurality of materials;
reconstructing an image according to the decomposed multiple material projection data to obtain a multi-material fusion image.
8. The decomposition system of claim 6, wherein said dual energy spectrum projection data comprises high energy spectrum projection data and low energy spectrum projection data;
the material template image comprises a high-density material template image and a low-density material template image;
the constructing a material template image from the dual-energy spectrum projection data comprises:
selecting the high-energy spectrum projection data or the low-energy spectrum projection data for image reconstruction to obtain a base image;
and dividing the base image to obtain a high-density material template image and a low-density material template image.
9. The decomposition system of claim 6, wherein said constructing a material decomposition image comprising a predetermined algebra from said dual spectral projection data and said polynomial comprises:
performing power operation according to the dual-energy spectrum projection data to obtain a plurality of power projection data;
respectively carrying out back projection reconstruction on the multiple power projection data to obtain an intermediate image;
and constructing a material decomposition image containing a preset algebra according to the intermediate image and the preset algebra.
10. The decomposition system of claim 6, wherein said solving for said values of said predetermined algebra from said material template image and said material decomposition image to obtain a plurality of target coefficients comprises:
constructing an evaluation function according to the material template image and the material decomposition image;
and solving the value of the preset algebra according to the evaluation function to obtain a plurality of target coefficients.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311211997.0A CN117252944A (en) | 2023-09-19 | 2023-09-19 | Decomposition method and system for dual-energy spectrum CT reconstructed image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311211997.0A CN117252944A (en) | 2023-09-19 | 2023-09-19 | Decomposition method and system for dual-energy spectrum CT reconstructed image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117252944A true CN117252944A (en) | 2023-12-19 |
Family
ID=89136288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311211997.0A Pending CN117252944A (en) | 2023-09-19 | 2023-09-19 | Decomposition method and system for dual-energy spectrum CT reconstructed image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117252944A (en) |
-
2023
- 2023-09-19 CN CN202311211997.0A patent/CN117252944A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9025815B2 (en) | System and method for multi-material correction of image data | |
| US7372934B2 (en) | Method for performing image reconstruction using hybrid computed tomography detectors | |
| US9036879B2 (en) | Multi-material decomposition using dual energy computed tomography | |
| US8774355B2 (en) | Method and apparatus for direct reconstruction in tomosynthesis imaging | |
| Maaß et al. | Exact dual energy material decomposition from inconsistent rays (MDIR) | |
| US8340373B2 (en) | Quantitative image reconstruction method and system | |
| US9683948B2 (en) | Systems and methods for iterative multi-material correction of image data | |
| US11207041B2 (en) | X-ray CT system and medical processing apparatus | |
| CN106233335A (en) | X-ray spectrum imaging | |
| US9374565B2 (en) | System and method for multi-material correction of image data | |
| US11607187B2 (en) | System and method for low-dose multi-spectral X-ray tomography | |
| CN102697516B (en) | Method for establishing fault contrastographic image display and computer fault contrastographic system | |
| JP2010125332A (en) | Sequential projection for forming calculator type tomographic image in optional spectrum | |
| US20040105528A1 (en) | Method and system for tomosynthesis image enhancement using transverse filtering | |
| CN111627083B (en) | Bone hardening artifact correction method, device, computer equipment and readable storage medium | |
| JP2020512900A (en) | Method of artifact reduction using single energy data in computed tomography | |
| Lu et al. | A scatter correction method for contrast-enhanced dual-energy digital breast tomosynthesis | |
| CN111818851B (en) | Non-spectral Computed Tomography (CT) scanner configured to generate spectral volumetric image data | |
| US11049295B2 (en) | Detection and/or correction of residual iodine artifacts in spectral computed tomography (CT) imaging | |
| CN110444276A (en) | Generate method, ct apparatus, program product and the data medium of image data | |
| US20190167218A1 (en) | Direct monochromatic image generation for spectral computed tomography | |
| CN117252944A (en) | Decomposition method and system for dual-energy spectrum CT reconstructed image | |
| US7924970B2 (en) | Method and device for generating a CT image with a high time resolution | |
| US9262825B2 (en) | Image reconstruction in interleaved multi-energy imaging | |
| US20210304458A1 (en) | Method for providing automatic adaptive energy setting for ct virtual momochromatic imaging |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |