Double-nuclide PET reconstruction method and device based on joint Poisson model

Abstract

The embodiment of the invention discloses a double-nuclide PET reconstruction method and device based on a joint Poisson model, and the method comprises the steps: coincidence data collection and random three-coincidence estimation: retaining the photon correlation in decay through a coincidence delay window, and estimating the random three-coincidence through the photon irrelevance between decay; establishing a combined Poisson distribution model, introducing impure electron nuclides, pure electron nuclides and random radioactive source distribution into the distribution model at the same time, and constructing a combined Poisson likelihood function; constructing a prior function according to the image features, and forming a target function with the combined Poisson likelihood function; and solving the maximum posteriori estimation of the target function to obtain the spatial distribution of non-pure electron nuclides, pure electron nuclides and random radioactive sources at the same time. The dual-nuclide PET imaging method can be directly applied to the dual-nuclide PET imaging field based on triple coincidence, and has the advantages of being capable of effectively inhibiting random triple coincidence artifacts and the like.

Description

Technical Field

[0001] The embodiments of the present disclosure belong to the technical field of medical image processing, and specifically relate to a dual-nuclide PET reconstruction method and device based on a joint Poisson model. Background Art

[0002] Dual-nuclide positron emission tomography (PET) is an advanced nuclear medicine imaging method that can inject two positron-emitting nuclides into patients at the same time, and achieve simultaneous imaging and separation of the dual nuclides through a single scan. Among them, the dual-nuclide PET imaging technology based on three-photon (prompt photon-annihilation photon pair) coincidence detection uses the differences in the number and energy of photons produced by different positron-emitting nuclides to directly distinguish the types of nuclides, and therefore has received widespread attention. Specifically, conventional PET uses a pair of annihilation photons with an energy of 511 keV produced by the decay of nuclides for two-photon coincidence imaging, while the above-mentioned dual-nuclide PET uses the physical property that impure positron-emitting nuclides simultaneously emit a pair of annihilation photons and a prompt photon with a certain probability, and detects two-photon coincidence and three-photon coincidence to distinguish impure positron-emitting nuclides from other positron-emitting nuclides.

[0003] Third, statistical characteristics, especially statistical counts and statistical accuracy, are the main factors affecting the imaging effect of dual-nuclide. Some impure positron nuclides commonly used in clinical practice, such as 68 Ga, the probability of generating instantaneous photons and triple coincidences is extremely low, and the triple coincidence statistical count cannot meet the requirements of quantitative imaging. In addition, the existence of random triple coincidences makes the three-photon coincidence no longer a unique property of impure positron nuclides, and the degradation of statistical accuracy also leads to the failure of dual-nuclide imaging. Prior to this, researchers have also conducted research on dual-nuclide imaging algorithms oriented to the statistical characteristics of triple coincidences. Andreyev and Celler et al. from the University of British Columbia proposed dual-nuclide EM reconstruction to address the problem of low triple coincidence statistical counts. This method considers the influence of impure positron nuclides and pure positron nuclides when constructing the objective function, and uses double coincidences as prior information for triple coincidence reconstruction to achieve dual-nuclide separation in iterations, but does not consider the impact of random triple coincidences on image quality, so dual-nuclide quantification cannot be achieved. Karp et al. from the University of Pennsylvania proposed an image domain post-processing method for the random triple coincidence problem. Through the linear correlation between the correlated random and double coincidences, the double coincidence image is used to correct the random triple coincidence in the triple coincidence image. However, this image domain method relies on the quality of the triple coincidence image and is limited by the triple coincidence statistical counts. Summary of the invention

[0004] The purpose of the embodiments of the present disclosure is to provide a dual-nuclide PET reconstruction method and device based on a joint Poisson model to solve the problems in the prior art.

[0005] In order to solve the above technical problems, the embodiments of the present disclosure adopt the following technical solutions:

[0006] In a first aspect, an embodiment of the present disclosure discloses a dual-nuclide PET reconstruction method based on a joint Poisson model, the method comprising:

[0007] Coincidence data collection and random triple coincidence estimation: preserve the intra-decay photon correlation through the consistent coincidence delay window, and estimate the random triple coincidence using the inter-decay photon irrelevance;

[0008] Establish a joint Poisson distribution model: introduce impure positron nuclides, pure positron nuclides and random radioactive source distribution into the distribution model at the same time, and construct a joint Poisson likelihood function;

[0009] Apply image features to establish priors: construct a prior function based on image features and form the objective function with the joint Poisson likelihood function;

[0010] Solving for the maximum a posteriori: Solve for the maximum a posteriori estimate of the objective function to simultaneously obtain the spatial distribution of impure positron nuclides, pure positron nuclides, and random radioactive sources.

[0011] In some embodiments, the coincidence data collection and random three-coincidence estimation include:

[0012] Coincidence data acquisition: Double coincidences consisting of 511keV photon pairs are extracted from single photon raw data using software coincidence method. p and a triple coincidence consisting of a 511keV photon pair and a prompt gamma p ;

[0013] Random triple coincidence estimation: Preserve the intra-decay photon correlation through the consistent coincidence delay window, and use the inter-decay photon irrelevance to estimate the random triple coincidence d .

[0014] In some embodiments, establishing a joint Poisson distribution model comprises:

[0015] Assuming a pure positron nuclide f A 、Impure positron nuclide f B and the random radiation source distribution f r Independently of each other, the process can be described as:

[0016]

[0017] in, The symbol indicates that the variable on the left side follows a probability distribution, dp A With dp Bdenote the true double coincidence events emitted by pure positron nuclides and the true double coincidence events emitted by impure positron nuclides, respectively. B represents a true triple coincidence event emitted by a non-pure positron nuclide, tp r represents a random triple coincidence event emitted by a random radiation source, td represents a delayed triple coincidence event extracted by a consistent coincidence delay window;

[0018] The impure positron nuclides, pure positron nuclides and random radioactive source distributions are simultaneously introduced into the distribution model to construct a joint Poisson likelihood function. The process can be described as:

[0019]

[0020] Among them, i and j represent the projection domain index and image domain index respectively, a ij represents the system response matrix, β represents the spatial variation normalization factor of the three-component system, ω A and ω B is a correction factor used to compensate for true double coincidences that constitute random triple coincidences but are not recorded.

[0021] In some embodiments, applying image features to establish priors includes:

[0022] According to the image characteristics of pure positron nuclides, impure positron nuclides and random radioactive sources, different differentiable prior functions P are constructed. A (f A ), P B (f B ) and P r (f r );

[0023] Different weights are assigned to the prior functions, which together with the joint Poisson likelihood function form the objective function F. The process can be described as:

[0024] F(f A ,f B ,f r )=logL(f A ,f B ,f r )+η A P A (f A )

[0025] +η B P B (f B )+η r P r (f r )

[0026] Among them, η A , ηB With η r They represent the weights of the prior functions of pure positron nuclides, impure positron nuclides and random radioactive sources respectively.

[0027] In some embodiments, solving the maximum a posteriori solution includes:

[0028] Using observed data to estimate complete data dp A (i,j), dp B (i,j),tp B (i,j),tp r The expectation of (i,j) and td(i,j) can be described as:

[0029]

[0030] Among them, double p Indicates instantaneous double coincidence observation value, triple p Indicates the instantaneous triple observation value, triple d represents the delayed three-compliance observation value;

[0031] The maximum a posteriori estimation is performed using the expectation maximization method, that is, the complete data expectation is substituted into the objective function, and the first-order partial derivative of the objective function is are zero respectively, and the distribution of binuclear nuclides when the objective function is maximized is obtained. The process can be described as:

[0032]

[0033]

[0034] Where k represents the current iteration number.

[0035] In some embodiments, the random three-component estimation includes:

[0036] The real-time single-photon raw data of different detectors are combined to obtain the overall single-photon data stream s p (n), n = 1, 2, ..., N, where N represents the total single photon counts;

[0037] The whole single photon raw data is subjected to a consistent time delay t * , generate delay data s d (n), n=1,2,…,N;

[0038] Will s p (n) and s d (n) Merge and sort by the arrival time of photons to generate a time-ordered single photon sequence s order (n), n=1,2,…,2N;

[0039] Using 511keV annihilation photon energy window and prompt photon energy window to study the order (n) Perform energy screening to obtain the photon sequence s filter (m), m = 1, 2, ..., M, where M represents the single photon count that meets the annihilation photon energy threshold or the prompt photon energy threshold;

[0040] For filter Each single photon event in (m) is subjected to a random triple coincidence judgment, and the triple coincidence event that meets the conditions is recorded as a delayed triple coincidence. d , the process can be described as: if s filter The three consecutive single photons in (m) all meet the time window, and only one of the three photons meets the instantaneous photon energy threshold, and at least one of the three photons comes from the immediate event s p (n), at least one single photon comes from the delayed event s d (n), then the triple event is recorded as delayed triple d .

[0041] In a second aspect, an embodiment of the present disclosure discloses a dual-nuclide PET reconstruction device based on a joint Poisson model, the device comprising:

[0042] The random triple coincidence estimation module is used to estimate the random triple coincidence count: a consistent delay window is used to delay single photons, the intra-decay photon correlation is retained, and the delayed triple coincidence is calculated using the inter-decay photon irrelevance, thereby estimating the random triple coincidence;

[0043] A joint Poisson data fidelity term building module is used to obtain the data fidelity term of the objective function: the impure positron nuclides, pure positron nuclides and random radioactive source distributions are simultaneously introduced into the distribution model to construct a joint Poisson likelihood function;

[0044] Image prior regularization term establishment module, used to obtain the regularization term of the objective function;

[0045] The solution module is used to solve the maximum a posteriori estimate of the objective function to simultaneously obtain the spatial distribution of impure positron nuclides, pure positron nuclides and random radioactive sources.

[0046] In a third aspect, an embodiment of the present disclosure discloses an electronic device, including:

[0047] one or more processors;

[0048] The storage unit is used to store one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement the dual-nuclide PET reconstruction method based on the joint Poisson model.

[0049] In a fourth aspect, an embodiment of the present disclosure discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the above-mentioned dual-nuclide PET reconstruction method based on a joint Poisson model.

[0050] In a fifth aspect, an embodiment of the present disclosure discloses a computer program product, including a computer program, which, when executed by a processor, can implement the above-mentioned dual-nuclide PET reconstruction method based on the joint Poisson model.

[0051] The dual-nuclide PET reconstruction method and device based on the joint Poisson model of the disclosed embodiment can be directly applied to the field of dual-nuclide PET imaging based on triple coincidence. In particular, for dual-nuclide imaging with low triple coincidence abundance and severe random triple coincidence contamination (e.g. 68 Ga / 18 F dual-nuclide imaging), the present invention constructs a joint Poisson model based on pure, impure positron nuclides and random radioactive sources, which can effectively suppress the influence of random triple coincidence artifacts while performing triple coincidence low count reconstruction, promoting the development of dual-nuclide quantitative imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0053] Figure 1 It is a flowchart of a dual-nuclide PET reconstruction method based on a joint Poisson model according to an embodiment of the present disclosure;

[0054] Figure 2 The ideal original image of the dual-nuclide phantom that needs to be scanned and reconstructed in the embodiment of the present disclosure;

[0055] Figure 3 This is a schematic diagram of the reconstruction results of the dual-nuclide phantom by conventional double coincidence PET;

[0056] Figure 4 This is a schematic diagram of the reconstruction results of the binuclear EM reconstruction algorithm for the binuclear phantom;

[0057] Figure 5 This is a schematic diagram of the reconstruction result of a dual-nuclide phantom according to an embodiment of the present invention;

[0058] Figure 6 It is a schematic structural diagram of a dual-nuclide PET reconstruction device based on a joint Poisson model according to another embodiment of the present disclosure.

[0059] Figure 7 The figure is a schematic diagram of the physical structure of an electronic device according to another embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] Various aspects and features of the present disclosure are described herein with reference to the drawings.

[0061] It should be understood that various modifications may be made to the embodiments of the present application. Therefore, the above description should not be considered as limiting, but only as an example of an embodiment. Other modifications within the scope and spirit of the present disclosure will occur to those skilled in the art.

[0062] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the general description of the present disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the present disclosure.

[0063] These and other characteristics of the present disclosure will become apparent from the following description of a preferred form of embodiment given as a non-limiting example, with reference to the attached drawings.

[0064] It should also be understood that, although the present disclosure has been described with reference to some specific examples, those skilled in the art will be able to realize many other equivalent forms of the present disclosure that have the features described in the claims and are therefore within the scope of protection defined thereby.

[0065] The above and other aspects, features and advantages of the present disclosure will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings.

[0066] Specific embodiments of the present disclosure are described hereinafter with reference to the accompanying drawings; however, it should be understood that the embodiments claimed are merely examples of the present disclosure, which may be implemented in a variety of ways. Well-known and / or repeated functions and structures are not described in detail to avoid obscuring the present disclosure with unnecessary or redundant details. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as a basis and representative basis for the claims to teach those skilled in the art to use the present disclosure in a variety of ways with substantially any suitable detailed structure.

[0067] This specification may use the phrases "in one embodiment," "in another embodiment," "in yet another embodiment," or "in other embodiments," all of which may refer to one or more of the same or different embodiments according to the present disclosure.

[0068] The purpose of the disclosed embodiment is to address the shortcomings of the existing dual-nuclide simultaneous imaging algorithm and propose a new dual-nuclide reconstruction method, which estimates the random triple coincidence through a consistent delay window, and introduces the random radioactive source, pure positron nuclide and impure positron nuclide into the objective function at the same time, suppresses the influence of the random triple coincidence and performs triple coincidence low count reconstruction, making dual-nuclide quantitative imaging possible.

[0069] In one embodiment, if Figure 1 As shown, an embodiment of the present disclosure provides a dual-nuclide PET reconstruction method S100 based on a joint Poisson model, and the method S100 includes the following steps S101 to S105.

[0070] Step S101, coincidence data collection and random three-coincidence estimation, retaining the intra-decay photon correlation through a consistent coincidence delay window, and estimating the random three-coincidence using the inter-decay photon irrelevance.

[0071] Specifically, in this step, the instant single-photon raw data of different detectors are first merged to obtain the overall single-photon data stream s p (n), n = 1, 2, ..., N, where N represents the total single photon counts;

[0072] Then, the overall single-photon raw data is subjected to a consistent time delay t * , generate delay data s d (n), n=1,2,…,N;

[0073] Then, s p (n) and s d (n) Merge and sort by the arrival time of photons to generate a time-ordered single photon sequence s order (n), n=1,2,…,2N;

[0074] Then, the 511keV annihilation photon energy window and the prompt photon energy window were used to order (n) Perform energy screening to obtain the photon sequence s filter (m), m = 1, 2, ..., M, where M represents the single photon count that meets the annihilation photon energy threshold or the prompt photon energy threshold;

[0075] Finally, for s filter Each single photon event in (m) is subjected to a random triple coincidence judgment, and the triple coincidence event that meets the conditions is recorded as a delayed triple coincidence. d Specifically, the process can be described as follows: if s filter The three consecutive single photons in (m) all meet the time window, and only one of the three photons meets the instantaneous photon energy threshold, and at least one of the three photons comes from the immediate event s p(n), at least one single photon comes from the delayed event s d (n), then the triple event is recorded as delayed triple d .

[0076] Step S102: Establish a joint Poisson distribution model, introduce the impure positron nuclides, pure positron nuclides and random radioactive source distribution into the distribution model at the same time, and construct a joint Poisson likelihood function.

[0077] Specifically, in this step, we first assume that the pure positron nuclide f A 、Impure positron nuclide f B and the random radiation source distribution f r Independently of each other, the process can be described as:

[0078]

[0079] In formula (1), The symbol indicates that the variable on the left side follows a certain probability distribution, dp A With dp B denote the true double coincidence events emitted by pure positron nuclides and the true double coincidence events emitted by impure positron nuclides, respectively. B represents a true triple coincidence event emitted by a non-pure positron nuclide, tp r represents a random triple coincidence event emitted by a random radiation source, td represents a delayed triple coincidence event extracted by a consistent coincidence delay window;

[0080] Then, the impure positron nuclides, pure positron nuclides and random radioactive source distributions are simultaneously introduced into the distribution model to construct a joint Poisson likelihood function. The process can be described as:

[0081]

[0082] In formula (2), i and j represent the projection domain index and image domain index respectively, a ij represents the system response matrix, β represents the spatial variation normalization factor of the three-component system, ω A and ω B is a correction factor used to compensate for true double coincidences that constitute random triple coincidences but are not recorded.

[0083] Step S103: construct a priori function according to the image features, and form an objective function with the joint Poisson likelihood function.

[0084] Specifically, in this step, different differentiable prior functions P are first constructed according to the image characteristics of pure positron nuclides, impure positron nuclides, and random radioactive sources. A (f A ), P B (fB ) and P r (f r );

[0085] Then, different weights are assigned to the prior functions, which together with the joint Poisson likelihood function form the objective function F. Specifically, the process can be described as:

[0086]

[0087] In formula (3), η A , η B With η r They represent the weights of the prior functions of pure positron nuclides, impure positron nuclides and random radioactive sources respectively.

[0088] Step S104, solving the maximum a posteriori estimate of the objective function to simultaneously obtain the spatial distribution of impure positron nuclides, pure positron nuclides and random radioactive sources.

[0089] Specifically, in this step, the complete data dp is first estimated using the observed data A (i,j), dp B (i,j),tp B (i,j),tp r (i, j) and td(i, j). Specifically, the process can be described as:

[0090]

[0091] In formula (4), double p Indicates instantaneous double coincidence observation value, triple p Indicates the instantaneous triple observation value, triple d represents the delayed three-compliance observation value;

[0092] Then, the maximum a posteriori estimation is performed using the expectation maximization method, that is, the complete data expectation is substituted into the objective function, and the first-order partial derivative of the objective function is are zero respectively, and the distribution of binuclear nuclides when the objective function is maximized is obtained. Specifically, the process can be described as:

[0093]

[0094] In formula (5), k represents the current iteration number.

[0095] In order to verify the dual-nuclide PET reconstruction method based on the joint Poisson model of this embodiment, Figure 2 to Figure 5 As shown, Figure 2 This is the ideal original image of the dual-nuclide phantom that needs to be scanned and reconstructed. The radioactive sources No. 1 to No. 3 in the image also include 18 F and68 Ga, radioactive sources 4 to 6 only contain 68 Ga is a nuclide, and radioactive sources 7 to 9 only contain 18 F is a nuclide, Figure 3 This is a schematic diagram of the reconstruction results of the dual-nuclide phantom by conventional double coincidence PET. 18 F and 68 Ga two nuclides, Figure 4 This is a schematic diagram of the reconstruction results of the dual-nuclide EM reconstruction algorithm for the dual-nuclide phantom, which can be distinguished 18 F nuclide, but because the influence of random triple coincidence is not considered, 68 The Ga separation diagram does not correspond to the ideal situation. Figure 5 The reconstruction result of the embodiment of the present invention is shown in FIG. 1 . The artifacts of the random three-dimensional image are suppressed, and the reconstructed 18 F and 68 Ga and Figure 2 The ideal image shown is consistent.

[0096] The dual-nuclide PET reconstruction method based on the joint Poisson model of the disclosed embodiment can be directly applied to the field of dual-nuclide PET imaging based on triple coincidence, especially for dual-nuclide imaging with low triple coincidence abundance and severe random triple coincidence contamination (e.g. 68 Ga / 18 F dual-nuclide imaging), the disclosed embodiment promotes the development of dual-nuclide quantitative imaging by constructing a joint Poisson model based on pure and impure positron nuclides and random radioactive sources, suppressing the influence of random triple coincidences while performing triple coincidence low count reconstruction.

[0097] Specifically, the dual-nuclide PET reconstruction method based on the joint Poisson model of the embodiment of the present disclosure has the following advantages:

[0098] Reduce artifact interference: By constructing a joint Poisson model, the image artifacts caused by random triple coincidences are effectively suppressed, improving the clarity and accuracy of imaging.

[0099] Improved dual-nuclide differentiation capability: By introducing the distribution of impure positron nuclides, pure positron nuclides, and random radioactive sources into the same model, different nuclides can be separated more accurately, achieving higher-precision dual-nuclide imaging.

[0100] Optimized low count reconstruction: High-quality reconstructed images can still be obtained when the triple coincidence counts are low, which is particularly suitable for imaging applications of rare nuclides and promotes the development of dual-nuclide quantitative imaging.

[0101] Enhanced quantitative imaging accuracy: Image features are used to construct a priori functions, combined with maximum a posteriori estimation, so that the imaging system has higher quantitative analysis capabilities and provides more reliable data support for clinical diagnosis.

[0102] Based on the same inventive concept, Figure 6 As shown, the embodiment of the present disclosure also provides a dual-nuclide PET reconstruction device 700 based on a joint Poisson model, which can be applied to the method described above. For details, please refer to the relevant records in the previous text, which will not be repeated here. The device 700 includes a random three-coincidence estimation module 701, a joint Poisson data fidelity term establishment module 702, an image prior regularization term establishment module 703 and a solution module 704.

[0103] The random triple coincidence estimation module 701 is used to estimate the random triple coincidence count, delay the single photon using a consistent delay window, retain the photon correlation within the decay, and use the photon irrelevance between decays to calculate the delayed triple coincidence, thereby estimating the random triple coincidence. The joint Poisson data fidelity term establishment module 702 is used to obtain the data fidelity term of the objective function, and introduce the distribution of impure positron nuclides, pure positron nuclides and random radioactive sources into the distribution model at the same time, thereby constructing a joint Poisson likelihood function. The image prior regularization term establishment module 703 is used to obtain the regularization term of the objective function. The solution module 704 is used to solve the maximum a posteriori estimate of the objective function to simultaneously obtain the spatial distribution of impure positron nuclides, pure positron nuclides and random radioactive sources.

[0104] The dual-nuclide PET reconstruction device based on the joint Poisson model of the disclosed embodiment can be directly applied to the field of dual-nuclide PET imaging based on triple coincidence, especially for dual-nuclide imaging with low triple coincidence abundance and severe random triple coincidence contamination (e.g. 68 Ga / 18 F dual-nuclide imaging), the disclosed embodiment promotes the development of dual-nuclide quantitative imaging by constructing a joint Poisson model based on pure and impure positron nuclides and random radioactive sources, suppressing the influence of random triple coincidences while performing triple coincidence low count reconstruction.

[0105] Based on the same inventive concept, the embodiment of the present disclosure also provides an electronic device for dual-nuclide PET reconstruction based on a joint Poisson model. Figure 7The physical structure diagram of the electronic device provided in the embodiment of the present invention may include: a processor 301, a communication interface 302, a memory 303 and a bus 304, wherein the processor 301, the communication interface 302 and the memory 303 communicate with each other through the bus 304. The processor 301 may call a computer program stored in the memory 303 and executable on the processor 301 to execute the dual-nuclide PET reconstruction method based on the joint Poisson model provided in the above embodiment. For example, it includes: coincidence data collection and random triple coincidence estimation: retaining the photon correlation within the decay through the consistent coincidence delay window, and estimating the random triple coincidence using the photon irrelevance between decays; establishing a joint Poisson distribution model: introducing the distribution of impure positron nuclides, pure positron nuclides and random radioactive sources into the distribution model at the same time, and constructing a joint Poisson likelihood function; applying image features to establish priors: constructing a priori functions based on image features, and forming the objective function with the joint Poisson likelihood function; solving the maximum a posteriori: solving the maximum a posteriori estimate of the objective function to simultaneously obtain the spatial distribution of impure positron nuclides, pure positron nuclides and random radioactive sources.

[0106] In addition, the logic instructions in the above-mentioned memory 303 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the embodiment of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0107] Based on the same inventive concept, an embodiment of the present disclosure further provides a computer program product, including a computer program, which, when executed by a processor, can implement the dual-nuclide PET reconstruction method based on the joint Poisson model as described above.

[0108] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0109] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0110] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

[0111] Multiple embodiments of the present disclosure are described in detail above, but the present disclosure is not limited to these specific embodiments. Those skilled in the art can make various variations and modifications to the embodiments based on the concepts of the present disclosure, and these variations and modifications should all fall within the scope of protection required by the present disclosure.

Claims

1. A dual-nuclide PET reconstruction method based on a joint Poisson model, characterized in that: The method comprises: Coincidence data collection and random triple coincidence estimation: preserve the intra-decay photon correlation through the consistent coincidence delay window, and estimate the random triple coincidence using the inter-decay photon irrelevance; Establish a joint Poisson distribution model: introduce impure positron nuclides, pure positron nuclides and random radioactive source distribution into the distribution model at the same time, and construct a joint Poisson likelihood function; Apply image features to establish priors: construct a prior function based on image features and form the objective function with the joint Poisson likelihood function; Solving for the maximum a posteriori: Solve for the maximum a posteriori estimate of the objective function to simultaneously obtain the spatial distribution of impure positron nuclides, pure positron nuclides, and random radioactive sources.

2. The dual-nuclide PET reconstruction method based on the joint Poisson model according to claim 1, characterized in that: The data collection and random three-component estimation include: Coincidence data acquisition: Double coincidences consisting of 511keV photon pairs are extracted from single photon raw data using software coincidence method. p and a triple coincidence consisting of a 511keV photon pair and a prompt gamma p ; Random triple coincidence estimation: Preserve the intra-decay photon correlation through the consistent coincidence delay window, and use the inter-decay photon irrelevance to estimate the random triple coincidence d .

3. The dual-nuclide PET reconstruction method based on the joint Poisson model according to claim 1, characterized in that: The establishing of the joint Poisson distribution model comprises: Assuming a pure positron nuclide f A 、Impure positron nuclide f B and the random radiation source distribution f r Independently of each other, the process can be described as: in, The symbol indicates that the variable on the left side follows a probability distribution, dp A With dp B denote the true double coincidence events emitted by pure positron nuclides and the true double coincidence events emitted by impure positron nuclides, respectively. B represents a true triple coincidence event emitted by a non-pure positron nuclide, tp r represents a random triple coincidence event emitted by a random radiation source, td represents a delayed triple coincidence event extracted by a consistent coincidence delay window; The impure positron nuclides, pure positron nuclides and random radioactive source distributions are simultaneously introduced into the distribution model to construct a joint Poisson likelihood function. The process can be described as: Among them, i and j represent the projection domain index and image domain index respectively, a ij represents the system response matrix, β represents the spatial variation normalization factor of the three-component system, ω A and ω B is a correction factor used to compensate for true double coincidences that constitute random triple coincidences but are not recorded.

4. The dual-nuclide PET reconstruction method based on the joint Poisson model according to claim 1, characterized in that: The applying image features to establish priors includes: According to the image characteristics of pure positron nuclides, impure positron nuclides and random radioactive sources, different differentiable prior functions P are constructed. A (f A ), P B (f B ) and P r (f r ); Different weights are assigned to the prior functions, which together with the joint Poisson likelihood function form the objective function F. The process can be described as: F(f A ,f B ,f r )=logL(f A ,f B ,f r )+η A P A (f A ) +η B P B (f B )+η r P r (f r ) Among them, η A , η B With η r They represent the weights of the prior functions of pure positron nuclides, impure positron nuclides and random radioactive sources respectively.

5. The dual-nuclide PET reconstruction method based on the joint Poisson model according to claim 1, characterized in that: The method of solving the maximum a posteriori solution includes: Using observed data to estimate complete data dp A (i,j), dp B (i,j),tp B (i,j),tp r The expectation of (i,j) and td(i,j) can be described as: Among them, double p Indicates instantaneous double coincidence observation value, triple p Indicates the instantaneous triple observation value, triple d represents the delayed three-compliance observation value; The maximum a posteriori estimation is performed using the expectation maximization method, that is, the complete data expectation is substituted into the objective function, and the first-order partial derivative of the objective function is are zero respectively, and the distribution of binuclear nuclides when the objective function is maximized is obtained. The process can be described as: Where k represents the current iteration number.

6. The dual-nuclide PET reconstruction method based on the joint Poisson model according to claim 2, characterized in that: The random three-component estimation includes: The real-time single-photon raw data of different detectors are combined to obtain the overall single-photon data stream s p (n), n = 1, 2, ..., N, where N represents the total single photon counts; The whole single photon raw data is subjected to a consistent time delay t * , generate delay data s d (n), n=1,2,…,N; Will s p (n) and s d (n) Merge and sort by the arrival time of photons to generate a time-ordered single photon sequence s order (n), n=1,2,…,2N; Using 511keV annihilation photon energy window and prompt photon energy window to study the order (n) Perform energy screening to obtain the photon sequence s filter (m), m = 1, 2, ..., M, where M represents the single photon count that meets the annihilation photon energy threshold or the prompt photon energy threshold; For filter Each single photon event in (m) is subjected to a random triple coincidence judgment, and the triple coincidence event that meets the conditions is recorded as a delayed triple coincidence. d , the process can be described as: if s filter The three consecutive single photons in (m) all meet the time window, and only one of the three photons meets the instantaneous photon energy threshold, and at least one of the three photons comes from the immediate event s p (n), at least one single photon comes from the delayed event s d (n), then the triple event is recorded as delayed triple d .

7. A dual-nuclide PET reconstruction device based on a joint Poisson model, characterized in that: The device comprises: The random triple coincidence estimation module is used to estimate the random triple coincidence count: a consistent delay window is used to delay single photons, the intra-decay photon correlation is retained, and the delayed triple coincidence is calculated using the inter-decay photon irrelevance, thereby estimating the random triple coincidence; A joint Poisson data fidelity term building module is used to obtain the data fidelity term of the objective function: the impure positron nuclides, pure positron nuclides and random radioactive source distributions are simultaneously introduced into the distribution model to construct a joint Poisson likelihood function; Image prior regularization term establishment module, used to obtain the regularization term of the objective function; The solution module is used to solve the maximum a posteriori estimate of the objective function to simultaneously obtain the spatial distribution of impure positron nuclides, pure positron nuclides and random radioactive sources.

8. An electronic device, characterized in that: include: one or more processors; A storage unit, used to store one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 can be implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 can be implemented.

Images