Scattering correction method and device, digital device and computer readable storage medium

Abstract

The application provides a scattering correction method and device, a digital device and a computer readable storage medium. The method comprises the following steps: obtaining a corresponding relationship function of one-dimensional detection energy spectrum and one-dimensional gamma photon energy spectrum; based on the corresponding relationship function, using a maximum likelihood-expectation maximization iterative algorithm to obtain two one-dimensional gamma photon energy spectrums corresponding to each down-sampling response line; based on the one-dimensional gamma photon energy spectrums, obtaining the number of non-scattering single events corresponding to each down-sampling response line; based on the total number of single events and the number of non-scattering single events corresponding to each down-sampling response line, calculating the number of scattering coincidence events corresponding to each down-sampling response line; performing up-sampling processing on the down-sampling response line to obtain an up-sampling response line; and calculating the number of scattering coincidence events corresponding to each up-sampling response line. When the scattering correction is performed, the application has the advantages of simple implementation and high accuracy.

Description

Technical Field

[0001] This application relates to the field of data processing, and more specifically, to a scattering correction method, apparatus, digitizing device, and computer-readable storage medium. Background Technology

[0002] Positron emission tomography (PET) works by labeling a compound with a radionuclide that can participate in blood flow or metabolic processes in living tissue. This compound is then injected into the organism. The positrons emitted by the radionuclide combine with electrons in the organism, resulting in electron-electron annihilation and producing two gamma photons of equal energy but opposite directions. Because these two gamma photons travel in different directions, the detector detects them at different times. If two scintillation crystals on the Line of Response (LOR) in the detector detect both gamma photons within a specified coincidence time window (e.g., 0–15 nanoseconds), the event of detecting these two gamma photons is called a coincidence event.

[0003] Coincidence events generally include true coincidence events, scattering coincidence events, and random coincidence events. A true coincidence event occurs when the time difference between the arrival of two gamma photons from the same annihilation event at two scintillation crystals located on the response line falls within the coincidence time window. A random coincidence event is a false coincidence event; in a random coincidence event, two detected gamma photons originate from different annihilation events but are mistakenly identified as occurring "simultaneously" within the coincidence time window. A scattering coincidence event refers to the following: for two gamma photons detected from the same annihilation event, one of the gamma photons changes its flight direction due to physical effects such as Compton scattering and / or Rayleigh scattering during its flight.

[0004] Of these three types of coincidence events, the data acquired for random coincidence events and scattering coincidence events may be erroneous, which could affect the resolution, contrast, and positioning accuracy of PET imaging. Therefore, correcting the acquired coincidence events is crucial.

[0005] Currently, commonly used scattering correction methods mainly include multi-window techniques, convolution / deconvolution techniques, and simulation-based techniques. Among these, simulation-based techniques are the most accurate and widely used.

[0006] Simulation-based techniques generally include single-scattering simulation, two-scattering simulation, and Monte Carlo simulation. Single-scattering simulation, for each response line, selects a scattering point from the input activity and decay images to obtain the photon path of a single scattering (i.e., a pair of gamma photons scatters only once). It calculates the single-scattering events along all photon paths corresponding to that response line. Finally, using tail data containing only scattering events, it fits the obtained single-scattering events using tail fitting (TF) to obtain a stretching factor for the scattering coincidence events, which is then applied to all data to achieve scattering correction. Two-scattering simulation is often used in conjunction with single-scattering simulation. The main difference is that two-scattering simulation uses two scattering points to determine the photon path. Monte Carlo simulation primarily determines the total scattering distribution by simulating the motion of each gamma photon pair determined based on the input activity and decay images.

[0007] In the process of developing this application, the inventors discovered that simulation-based technologies have at least the following problems:

[0008] (1) Single scattering simulation methods only consider the case of single scattering. However, in reality, multiple scattering may occur, which may lead to lower accuracy of scattering correction results and thus lower quality of reconstructed images. Although double scattering simulation methods and Monte Carlo simulation methods consider the case of multiple scattering, the calculation process is complex and computationally intensive, which slows down the image reconstruction process and is inefficient.

[0009] (2) Scattering events need to be determined before reconstructing the activity image. However, simulation-based methods for determining scattering events require prior knowledge of the activity image, and existing technologies make it difficult to obtain accurate activity images.

[0010] (3) Since it is difficult to obtain activity images outside the field of view, existing technologies usually do not take into account external radiation (i.e., scattering events generated by gamma photon pairs located outside the field of view), which may lead to lower accuracy of the correction results and thus affect the quality of the reconstructed image.

[0011] (4) The OSL-EDR and OT-EDR algorithms currently in use require the selection of appropriate hyperparameter β to achieve good results. The optimal δ needs to be adjusted through some preliminary experiments, and different sizes of prostheses require different β values, which brings many inconveniences to the use of OSL-EDR and OT-EDR algorithms.

[0012] Therefore, there is an urgent need to provide a scattering correction method that is different from simulation-based methods, does not depend on activity images and attenuation images, and does not require an appropriate hyperparameter β.

[0013] The content in the background section merely discloses technology known only to the inventors and is not intended to represent prior art in the field. Summary of the Invention

[0014] This application aims to provide a scattering correction method, apparatus, digitization device, and computer-readable storage medium to solve at least one problem existing in the prior art.

[0015] According to a first aspect of this application, a scattering correction method is provided, the scattering correction method comprising: obtaining a correspondence function between a one-dimensional detector energy spectrum and a one-dimensional gamma photon energy spectrum; based on the correspondence function, using a maximum likelihood-expectation-maximization iterative algorithm and initial iterative values ​​corresponding to downsampled response lines to obtain two one-dimensional gamma photon energy spectra corresponding to each downsampled response line; obtaining the number of unscattered single events corresponding to each downsampled response line based on the one-dimensional gamma photon energy spectrum; calculating the number of scattering coincidence events corresponding to each downsampled response line based on the total number of single events and the number of unscattered single events corresponding to each downsampled response line; performing upsampling processing on the downsampled response lines to obtain upsampled response lines; and calculating the number of scattering coincidence events corresponding to each upsampled response line.

[0016] In some embodiments, before obtaining the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum, the method further includes: obtaining downsampled response lines based on the detector data; and obtaining two one-dimensional detector energy spectra based on each downsampled response line.

[0017] In some embodiments, the downsampled response line is acquired based on probe data, including: acquiring a coincident response line based on the probe data; and downsampling the coincident response line to obtain a downsampled response line; wherein the probe data is acquired based on a target object.

[0018] In some embodiments, the one-dimensional probe energy spectrum is obtained based on each downsampled response line, including:

[0019] Break down all coincidence events corresponding to the downsampling response line into single events;

[0020] Based on the location information of a single event, the single event is divided into two parts, and each part corresponds to a one-dimensional detection energy spectrum.

[0021] Based on the two parts of single events, two one-dimensional probe energy spectra are obtained.

[0022] In some embodiments, the location information of a single event includes the location information of the downsampling detection module that detected the single event.

[0023] In some embodiments, the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum is:

[0024]

[0025] Where p represents the fuzzy response matrix, Let X represent the expected value of the one-dimensional probe energy spectrum; let X represent the one-dimensional gamma photon energy spectrum; and let r represent the one-dimensional delayed energy spectrum corresponding to the random coincidence event.

[0026] In some embodiments, the one-dimensional delayed energy spectrum corresponding to the random coincidence event is obtained through delayed coincidence event estimation.

[0027] In some embodiments, obtaining a one-dimensional delayed energy spectrum includes: obtaining all delayed coincidence events in the downsampled response line based on a delayed coincidence window; and splitting the delayed coincidence events into delayed single events.

[0028] Based on the location information of the delayed single event, the delayed single event is divided into two parts, each part corresponding to a one-dimensional delayed energy spectrum;

[0029] Based on the two-part delayed single event, two one-dimensional delayed energy spectra are obtained;

[0030] The location information of the delayed single event includes the location information of the downsampling detection module that detected the delayed single event.

[0031] In some embodiments, obtaining the fuzzy response matrix p includes: obtaining prior probe data based on the spur; obtaining a prior downsampled response line based on the prior probe data; dividing all prior coincidence events corresponding to the prior downsampled response line into prior instant coincidence events and prior delayed coincidence events; obtaining a prior one-dimensional instantaneous energy spectrum based on the prior instantaneous coincidence events corresponding to the downsampled response line; obtaining a prior one-dimensional delayed energy spectrum based on the prior delayed coincidence events corresponding to the downsampled response line; obtaining the difference between the prior one-dimensional instantaneous energy spectrum and the prior one-dimensional delayed energy spectrum, normalizing the difference to obtain a one-dimensional intermediate energy spectrum; and obtaining the fuzzy response matrix based on the one-dimensional intermediate energy spectrum.

[0032] In some embodiments, obtaining the two one-dimensional gamma photon spectra corresponding to each downsampled response line includes: obtaining the maximum likelihood-expectation-maximization iterative algorithm formula based on the correspondence function between the one-dimensional probe energy spectrum and the one-dimensional gamma photon energy spectrum; setting the number of iterations, iterating based on the obtained initial iteration value using the maximum likelihood-expectation-maximization iterative algorithm formula until the set number of iterations is reached, and obtaining the one-dimensional gamma photon energy spectrum corresponding to the one-dimensional probe energy spectrum.

[0033] In some embodiments, the iterative formula of the maximum likelihood-expectation maximization iterative algorithm is:

[0034]

[0035] Among them, Yi In the diagram, Y represents the one-dimensional detection energy spectrum obtained from scanning the target object corresponding to a single downsampling response line, i represents the i-th vertical bar region in the energy spectrum, b and j represent the b-th and j-th vertical bar regions in the energy spectrum, respectively, k is the iteration number, and p ij and p ib In this context, p represents the fuzzy response matrix corresponding to the downsampled response line. ij p represents the value in the i-th row and j-th column of the fuzzy response matrix corresponding to this downsampled response line. ib r represents the value in the i-th row and b-th column of the fuzzy response matrix corresponding to this downsampled response line. i This indicates the one-dimensional delayed energy spectrum corresponding to the random coincidence events included in the downsampled response line; The one-dimensional gamma photon energy spectrum corresponding to the j-th vertical bar region of the energy spectrum obtained in the (k+1)-th iteration; The one-dimensional gamma photon energy spectra corresponding to the b-th and j-th vertical regions of the energy spectrum obtained in the k-th iteration are respectively represented.

[0036] In some embodiments, obtaining the initial iteration value includes: obtaining the global initial scattered photon energy spectrum; performing deblurring on the global initial scattered photon energy spectrum to obtain the global intermediate initial iteration value; and stretching the global intermediate initial iteration value to obtain the initial iteration value corresponding to each downsampled response line.

[0037] In some embodiments, obtaining the global initial scattered photon energy spectrum includes: obtaining coincidence response lines based on detection data; dividing all coincidence events corresponding to all coincidence response lines into instantaneous coincidence events and random coincidence events; obtaining the global one-dimensional instantaneous energy spectrum based on instantaneous coincidence events; obtaining the global one-dimensional delayed energy spectrum based on delayed coincidence events; subtracting the global one-dimensional delayed energy spectrum corresponding to delayed coincidence events from the global one-dimensional instantaneous energy spectrum corresponding to instantaneous coincidence events on all response lines to obtain the global one-dimensional undelayed energy spectrum; and obtaining the objective function of the global initial scattered photon energy spectrum based on the global one-dimensional undelayed energy spectrum.

[0038] In some embodiments, the objective function of the global initial scattered photon energy spectrum is:

[0039]

[0040] Among them, S i In this context, S represents the global initial scattered photon energy spectrum; i represents the i-th vertical bar region in the energy spectrum; C i In this context, C represents the global one-dimensional undelayed energy spectrum, obtained based on the target object, and U... i In this context, U represents the global one-dimensional intermediate energy spectrum acquired by the entire PET system scan; the global one-dimensional intermediate energy spectrum is acquired based on the prosthesis; hel and heu are the lower and upper thresholds of the set high-energy window, respectively. The tensile coefficient is denoted as .

[0041] In some embodiments, obtain U i This includes: acquiring prior detection data based on the prosthesis; acquiring all prior coincidence response lines based on the prior detection data; classifying all prior coincidence events corresponding to all prior coincidence response lines into global prior instant coincidence events and global prior delayed coincidence events; acquiring the global prior one-dimensional instantaneous energy spectrum based on the global prior instantaneous coincidence events; acquiring the global prior one-dimensional delayed energy spectrum based on the global prior delayed coincidence events; and acquiring the difference U between the global prior instantaneous energy spectrum and the global prior delayed energy spectrum. i .

[0042] In some embodiments, the global initial scattered photon energy spectrum is deblurred to obtain a global intermediate initial iteration value, including: obtaining a deblurring iteration formula based on a maximum likelihood-expectation-maximization iteration algorithm; setting a prior initial iteration value and the number of iterations, and iterating through the deblurring iteration formula until the number of iterations is reached to obtain the scattered part of the global initial energy spectrum; estimating the unscattered part of the global initial energy spectrum based on the scattered part of the global initial energy spectrum; and obtaining a global intermediate initial iteration value based on the scattered part of the global initial energy spectrum and the unscattered part of the global initial energy spectrum.

[0043] In some embodiments, the deblurring iterative formula is:

[0044]

[0045] Among them, S i The S characterization in the diagram is based on the global one-dimensional probe energy spectrum obtained by scanning the target object. i represents the i-th vertical bar region in the energy spectrum, b and j represent the b-th and j-th vertical bar regions in the energy spectrum, respectively, k is the iteration number, and p... ij and p ib In this context, p represents the fuzzy response matrix corresponding to the downsampled response line. ij Let P be the value in the i-th row and j-th column of the global fuzzy response matrix. ib r represents the value in the i-th row and b-th column of the global fuzzy response matrix. i This represents the global one-dimensional delayed energy spectrum corresponding to all random coincidence events; The global one-dimensional gamma photon energy spectrum corresponding to the j-th vertical stripe region obtained in the (k+1)-th iteration; The global one-dimensional gamma photon energy spectra corresponding to the b-th and j-th vertical stripe regions obtained in the k-th iteration are respectively characterized;

[0046]

[0047] Among them, largeconstant represents a constant.

[0048] In some embodiments, the global fuzzy response matrix P is obtained. ij The process includes: acquiring prior detection data based on the spoof; acquiring all prior coincidence response lines based on the prior detection data; dividing all prior coincidence events corresponding to all prior coincidence response lines into global prior instant coincidence events and global prior delayed coincidence events; acquiring the global prior one-dimensional instantaneous energy spectrum based on the global prior instantaneous coincidence events; acquiring the global prior one-dimensional delayed energy spectrum based on the global prior delayed coincidence events; obtaining the difference between the global prior instantaneous energy spectrum and the global prior delayed energy spectrum; normalizing the difference to obtain the global prior intermediate energy spectrum; and acquiring the global fuzzy response matrix based on the global prior intermediate energy spectrum.

[0049] In some embodiments, the unscattered portion of the global initial energy spectrum is estimated using an estimation function, which is:

[0050] X j init =δ×sum(PX) init，sc )×κ,j=511keV

[0051] Where P is the global fuzzy response matrix; δ is the hyperparameter; and κ is the ratio of unscattered photons to scattered photons.

[0052]

[0053] Where sum() represents summation, and C and S correspond to C and S respectively. i S i .

[0054] In some embodiments, the representation function of the global intermediate initial iteration value is:

[0055]

[0056] Where δ is a hyperparameter, P is the global fuzzy response matrix, and κ is the ratio of unscattered photons to scattered photons.

[0057] In some embodiments, stretching the intermediate initial iteration values ​​to obtain the initial iteration values ​​corresponding to each downsampled response line includes:

[0058] The global intermediate initial iteration value is stretched using a stretching function to obtain the initial iteration value corresponding to each downsampled response line. The stretching function is:

[0059]

[0060] Among them, X 0 The initial iteration value is given by `sum()`, where `P` represents the summation function, `X` is the global fuzzy response matrix, and `X` is the summation value. initis the global intermediate initial iteration value; Y is the one-dimensional probe energy spectrum obtained based on the target object scan corresponding to a single downsampled response line; r represents the one-dimensional delayed energy spectrum corresponding to the random coincidence events included in a single response line.

[0061] In some embodiments, obtaining the number of unscattered events corresponding to each downsampled response line based on the one-dimensional gamma photon energy spectrum includes: using the count value corresponding to the energy of the gamma photons that have not been scattered in the one-dimensional gamma photon energy spectrum as the number of unscattered single events; and using the sum of the number of unscattered single events in the two one-dimensional gamma photon energy spectra corresponding to each downsampled response line as the number of unscattered events corresponding to the corresponding downsampled response line.

[0062] In some embodiments, the number of scattering coincidence events corresponding to each downsampled response line is calculated using the following formula:

[0063] SC = S tot -S unsc ,

[0064] or

[0065] or

[0066] Among them, S tot The total number of single events is twice the total number of coincident events corresponding to all response lines; S unsc SC represents the number of unscattered single events; SC represents the number of scattering coincidence events.

[0067] In some embodiments, upsampling the downsampled response lines includes: processing the downsampled response lines using a bilinear interpolation method, a 4D linear interpolation method, or a 5D linear interpolation method to obtain the upsampled response lines corresponding to each downsampled response line.

[0068] In some embodiments, when processing the downsampled response line using a 4D linear interpolation method, the number of scattering coincidence events corresponding to the upsampled response line is calculated using the following formula:

[0069]

[0070] in, The number of scattering coincidence events characterizing the upsampled response line composed of upsampled crystals i and j after 4D linear interpolation; S i This represents the set of downsampling center crystal, downsampling axial adjacent crystal, downsampling radial adjacent crystal, and downsampling relative crystal corresponding to upsampling crystal i; K and L represent the crystal numbers corresponding to the downsampling response lines; Characterize the number of scattering coincidence events on the downsampled response line composed of downsampled crystals K and L; This represents the weight of the downsampling crystal K relative to the upsampling crystal i; W represents the weight of the downsampling crystal L relative to the upsampling crystal j; tot The normalization factor is characterized as follows:

[0071]

[0072] Where N is the number of upsampled response lines contained in a downsampled response line.

[0073] In some embodiments, calculating the number of scattering coincidence events corresponding to each upsampled response line includes: calculating the ratio of the total number of scattering coincidence events on each downsampled response line to the total number of coincidence events; and calculating the number of scattering coincidence events corresponding to the corresponding upsampled response line based on the number of coincidence events on each upsampled response line included in each downsampled response line and the ratio.

[0074] According to a second aspect of this application, a scattering correction method is provided, the scattering correction method comprising: obtaining a correspondence function between a one-dimensional detector energy spectrum and a one-dimensional gamma photon energy spectrum; based on the correspondence function, using a maximum likelihood-expectation-maximization iterative algorithm and initial iterative values ​​corresponding to the response lines to obtain two one-dimensional gamma photon energy spectra corresponding to each response line; obtaining the number of unscattered single events corresponding to each response line based on the one-dimensional gamma photon energy spectrum; and calculating the number of scattering coincidence events corresponding to each response line based on the total number of single events corresponding to each response line and the number of unscattered single events.

[0075] In some embodiments, the one-dimensional gamma photon energy spectrum is obtained based on each response line, including: obtaining the maximum likelihood-expectation-maximization iterative algorithm formula based on the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum; setting the number of iterations, iterating based on the obtained initial iteration value using the maximum likelihood-expectation-maximization iterative algorithm formula until the set number of iterations is reached, and obtaining the one-dimensional gamma photon energy spectrum corresponding to the one-dimensional detector energy spectrum.

[0076] In some embodiments, obtaining the initial iteration value includes: obtaining the global initial scattered photon energy spectrum; performing deblurring on the global initial scattered photon energy spectrum to obtain the global intermediate initial iteration value; and stretching the global intermediate initial iteration value to obtain the initial iteration value corresponding to each response line.

[0077] In some embodiments, obtaining the global initial scattered photon energy spectrum includes: obtaining coincidence response lines based on detection data; dividing all coincidence events corresponding to all coincidence response lines into instantaneous coincidence events and random coincidence events; obtaining the global one-dimensional instantaneous energy spectrum based on instantaneous coincidence events; obtaining the global one-dimensional delayed energy spectrum based on delayed coincidence events; subtracting the global one-dimensional delayed energy spectrum corresponding to delayed coincidence events from the global one-dimensional instantaneous energy spectrum corresponding to instantaneous coincidence events on all response lines to obtain the global one-dimensional undelayed energy spectrum; and obtaining the objective function of the global initial scattered photon energy spectrum based on the global one-dimensional undelayed energy spectrum.

[0078] In some embodiments, the global initial scattered photon energy spectrum is deblurred to obtain a global intermediate initial iteration value, including: obtaining a deblurring iteration formula based on a maximum likelihood-expectation-maximization iteration algorithm; setting a prior initial iteration value and the number of iterations, and iterating through the deblurring iteration formula until the number of iterations is reached to obtain the scattered part of the global initial energy spectrum; estimating the unscattered part of the global initial energy spectrum based on the scattered part of the global initial energy spectrum; and obtaining a global intermediate initial iteration value based on the scattered part of the global initial energy spectrum and the unscattered part of the global initial energy spectrum.

[0079] In some embodiments, obtaining the number of unscattered events corresponding to each response line based on the one-dimensional gamma photon energy spectrum includes: using the count value corresponding to the energy of the gamma photons that have not been scattered in the one-dimensional gamma photon energy spectrum as the number of unscattered single events; and using the sum of the number of unscattered single events in the two one-dimensional gamma photon energy spectra corresponding to each response line as the number of unscattered events corresponding to the corresponding response line.

[0080] In some embodiments, the number of scattering coincidence events corresponding to each response line is calculated using the following formula:

[0081] SC = S tot -S unsc ,

[0082] or

[0083] or

[0084] Among them, S tot The total number of single events is twice the total number of coincident events corresponding to all response lines; S unsc SC represents the number of unscattered single events; SC represents the number of scattering coincidence events.

[0085] In some embodiments, before obtaining the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum, the method further includes: obtaining response lines based on the detector data; and obtaining two one-dimensional detector energy spectra based on each response line.

[0086] According to a third aspect of this application, an image reconstruction method is provided, the image reconstruction method comprising: obtaining the number of scattering coincidence events corresponding to each response line based on the scattering correction method described in any of the above embodiments; and reconstructing the image using an iterative image reconstruction algorithm based on the number of scattering coincidence events to obtain a reconstructed image.

[0087] According to a fourth aspect of this application, a scattering correction device is provided, the scattering correction device comprising: a correspondence acquisition module configured to acquire a correspondence function between a one-dimensional detector energy spectrum and a one-dimensional gamma photon energy spectrum; a gamma photon energy spectrum acquisition module configured to acquire two one-dimensional gamma photon energy spectra corresponding to each downsampled response line based on the correspondence function, using a maximum likelihood-expectation-maximization iterative algorithm and initial iterative values ​​corresponding to the downsampled response lines; an unscattered single event quantity acquisition module configured to acquire the quantity of unscattered single events corresponding to each downsampled response line based on the one-dimensional gamma photon energy spectrum; a scattering coincidence event quantity acquisition module configured to calculate the quantity of scattering coincidence events corresponding to each downsampled response line based on the total number of single events corresponding to each downsampled response line and the quantity of unscattered single events; an upsampling module configured to perform upsampling processing on the downsampled response lines to acquire upsampled response lines; and a scattering coincidence event quantity calculation module configured to calculate the quantity of scattering coincidence events corresponding to each upsampled response line.

[0088] According to a fifth aspect of this application, a scattering correction device is provided, the scattering correction device comprising: a correspondence acquisition module configured to acquire a correspondence function between a one-dimensional detector energy spectrum and a one-dimensional gamma photon energy spectrum; a gamma photon energy spectrum acquisition module configured to acquire two one-dimensional gamma photon energy spectra corresponding to each response line based on the correspondence function, using a maximum likelihood-expectation-maximization iterative algorithm and initial iterative values ​​corresponding to the response lines; an unscattered single event quantity acquisition module configured to acquire the quantity of unscattered single events corresponding to each response line based on the one-dimensional gamma photon energy spectrum; and a scattering coincidence event quantity acquisition module configured to calculate the quantity of scattering coincidence events corresponding to each response line based on the total number of single events corresponding to each response line and the quantity of unscattered single events.

[0089] According to a sixth aspect of this application, an image reconstruction apparatus is provided, the image reconstruction apparatus comprising: a scattering coincidence event number acquisition module configured to acquire the number of scattering coincidence events corresponding to each response line based on the scattering correction apparatus described in any of the above embodiments; and a reconstruction module configured to perform image reconstruction using an iterative image reconstruction algorithm based on the number of scattering coincidence events, and acquire a reconstructed image.

[0090] According to a seventh aspect of this application, a digital device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the scattering correction method as described in any of the above embodiments.

[0091] According to an eighth aspect of this application, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the scattering correction method described in any of the above embodiments.

[0092] Based on the above embodiments of this application, the beneficial effects of this application include: In the embodiments of this application, scattering correction is performed based on the maximum likelihood-expectation-maximization iterative algorithm (MLEM), which does not depend on the activity image and attenuation image, making scattering correction easy to implement. Moreover, this method does not require the hyperparameter β, so it is no longer necessary to set the hyperparameter β within a certain range, that is, it is minimally affected by the hyperparameter β and hardly needs to be adjusted, making scattering correction more convenient, simple to implement and highly accurate. Attached Figure Description

[0093] The embodiments of this application are described in detail below with reference to the accompanying drawings. The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0094] Figure 1 An exemplary flowchart of a scattering correction method according to an example embodiment of this application is shown;

[0095] Figure 2 This diagram illustrates the calculation of crystal weights in 4D linear interpolation according to an example embodiment of this application.

[0096] Figure 3 An exemplary flowchart of a scattering correction method according to another example embodiment of this application is shown;

[0097] Figure 4 An exemplary flowchart of an image reconstruction method according to an example embodiment of this application is shown;

[0098] Figure 5 An exemplary block diagram of a scattering correction method according to an example embodiment of this application is shown;

[0099] Figure 6 An exemplary block diagram of a scattering correction method according to another example embodiment of this application is shown;

[0100] Figure 7 An exemplary block diagram of an image reconstruction apparatus according to an example embodiment of this application is shown. Detailed Implementation

[0101] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of this application. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0102] In the description of this application, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used only for description and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more similar features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0103] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection, an electrical connection, or a communication connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0104] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature being directly above or diagonally above the second feature, or simply indicates that the relative height of the first feature in a certain dimension is higher than that of the second feature. "Below," "below," and "under" of the second feature includes the first feature being directly above or diagonally above the second feature, or simply indicates that the relative position of the first feature in a certain dimension is smaller than that of the second feature.

[0105] Different embodiments or examples are provided below to implement different structures of this application. To simplify this application, the components and arrangements of specific examples are described below. Of course, these are merely examples and are not intended to limit this application. Reference numerals may be repeated in different examples; such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements described. Furthermore, this application provides examples of various specific processes and materials, but those skilled in the art can apply other processes and / or substitute other materials based on the teachings of this application.

[0106] The following description, with reference to the accompanying drawings, illustrates some preferred embodiments of the present application. It should be noted that the following description is for illustrative purposes only and is not intended to limit the scope of protection of this application.

[0107] Figure 1 This is an exemplary flowchart of a scattering correction method according to some embodiments of this application, see [link to flowchart]. Figure 1 The energy screening method 100 may include the following steps S110-S160.

[0108] In this embodiment of the application, before step S110, steps S0110-S0120 are further included:

[0109] S0110, Obtain the downsampling response line based on the probe data.

[0110] In some embodiments of this application, step S0110 may further include:

[0111] S001, Obtain probe data based on the target object.

[0112] In some embodiments, the target object may be a living object, including but not limited to humans, animals, etc.

[0113] In some embodiments, the detection data may be sampling data obtained based on detector detection and subsequently on a multiple voltage threshold (MVT) method. For example, the detection data may include voltage threshold-time pairs. Specifically, the specific design of the detector can be referred to in the prior art, and will not be elaborated here.

[0114] In some embodiments of this application, step S0110 may further include the following steps:

[0115] S0111, Obtain the response line based on the probe data.

[0116] In some embodiments, the probe data is filtered based on time windows and energy windows to obtain coincident events, thereby obtaining coincident response lines. Specific details can be found in existing technologies, which will not be elaborated upon here.

[0117] S0112, perform downsampling processing on the conformal response line to obtain a downsampled response line.

[0118] Those skilled in the art will understand that a response line refers to the connection line between a pair of scintillation crystals in a PET system's detector module pair. Since the number of coincidence events on a single response line may be relatively small during normal scanning, it is difficult to obtain the corresponding energy spectrum based on the coincidence events corresponding to a single response line. Therefore, downsampling processing is required to obtain a downsampled response line. For example, a PET system's detector module pair originally contains 3×3=9 coincidence response lines. After downsampling these 9 response lines, a single downsampled response line is obtained, formed by merging these 9 response lines. Specifically, downsampling processing can merge the response lines included within a single detector module pair, or it can merge the response lines included in multiple adjacent detector module pairs (e.g., two axially adjacent, four axially and radially adjacent, etc.).

[0119] In some embodiments, downsampling processing includes merging multiple parallel or angularly aligned response lines that meet preset conditions into a single downsampling response line, but is not limited thereto. Specific details of downsampling processing can be found in existing technologies and will not be elaborated upon here.

[0120] S0120, two one-dimensional probe energy spectra are obtained based on each downsampling response line.

[0121] In some embodiments, the energy spectrum is characterized as a one-dimensional energy histogram, where the horizontal axis represents energy and the vertical axis represents the gamma photon count. The one-dimensional detection energy spectrum refers to the energy spectrum composed of the energies of gamma photons detected by the detector. It should be understood that the energy spectrum histogram includes multiple vertical bar regions, each of which is called a bin; specific details can be found in existing technologies and will not be elaborated upon here.

[0122] In some embodiments of this application, step S0120 includes:

[0123] S0121, split all coincidence events corresponding to the downsampling response line into single events.

[0124] It should be understood that a coincidence event contains two individual events. Each coincidence event corresponding to a downsampled response line is broken down into two corresponding individual events, and each individual event contains location information, energy information, and time information.

[0125] S0122, a single event is divided into two parts based on its location information, with each part corresponding to a one-dimensional detection energy spectrum.

[0126] In some embodiments, the location information of a single event includes the location information of the downsampling detection module that detected the single event. For example, a downsampling response line corresponds to two detection modules A and B, and a portion of the single event included in the downsampling response line corresponds to detection module A, while the other portion corresponds to detection module B.

[0127] S0123, based on the division of two parts of single events, obtain two one-dimensional probe energy spectra.

[0128] In some specific examples, based on the detection module information corresponding to a single event, two one-dimensional detection energy spectra corresponding to the downsampled response line can be obtained. Continuing with the example above, one one-dimensional detection energy spectrum can be obtained based on the single event corresponding to detection module A, and the other one-dimensional detection energy spectrum corresponding to the downsampled response line can be obtained based on the single event corresponding to detection module B.

[0129] Steps S110-S160 of this embodiment are as follows:

[0130] Step S110: Obtain the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum.

[0131] In some embodiments, the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum is:

[0132]

[0133] in, denoted as the expected value of the one-dimensional probe energy spectrum obtained from scanning the target object corresponding to a single downsampled response line; X represents the one-dimensional gamma photon energy spectrum; p represents the fuzzy response matrix corresponding to a single downsampled response line; and r represents the one-dimensional delayed energy spectrum corresponding to the random coincidence events included in a single response line.

[0134] In some embodiments, the one-dimensional delayed energy spectrum corresponding to a random coincidence event is obtained through delayed coincidence event estimation.

[0135] In some specific embodiments, corresponding to step S0120, obtaining the one-dimensional delayed energy spectrum includes:

[0136] S1111: Obtain all delayed coincidence events in the downsampled response line based on the delayed coincidence window; split each delayed coincidence event into two corresponding delayed single events.

[0137] In some embodiments, the delayed coincidence window includes a delayed coincidence time window. Specifically, when the signal is delayed, the corresponding coincidence event is called the delayed coincidence event, and the corresponding time window is called the delayed coincidence time window. The delayed coincidence time window is specifically obtained based on prior information, for example, it can be set to 2ns-10ns. For details on how the delayed coincidence event is obtained, please refer to existing technologies, which will not be elaborated here. It should be understood that a delayed coincidence event contains two delayed single events. Each delayed coincidence event included in a downsampled response line is split into two corresponding delayed single events, and each delayed single event contains location information, energy information, and time information.

[0138] S1112, based on the location information of the delayed single event, divide the delayed single event into two parts, each part corresponding to a one-dimensional delayed energy spectrum.

[0139] In some specific embodiments, corresponding to step S0122, the location information of the delayed single event includes the location information of the downsampling detection module that detected the delayed single event.

[0140] S1113, based on the two-part delayed single event, obtain two one-dimensional delayed energy spectra.

[0141] In some specific examples, based on the detection module information corresponding to the delayed single event, the two one-dimensional delayed energy spectra corresponding to the downsampling response line can be obtained.

[0142] In some specific embodiments, obtaining the fuzzy response matrix p includes:

[0143] S1121, obtaining prior detection data based on a prosthesis.

[0144] In some embodiments, the prosthesis may employ a point source or a line source, but is not limited thereto.

[0145] S1122, Obtain the a priori downsampling response line based on prior detection data.

[0146] In some embodiments, corresponding to step S0110, step S1122 includes:

[0147] S11221, Obtain the a priori coincidence response line based on prior detection data;

[0148] S11222, downsample the prior conforming response line to obtain the prior downsampled response line.

[0149] In some specific examples, how to obtain the prior coincidence response line and the prior downsampled response line can be referred to step S0110, which will not be elaborated here.

[0150] S1123, divide all prior coincidence events corresponding to the prior downsampling response line into prior immediate coincidence events and prior delayed coincidence events.

[0151] In some embodiments, prior instantaneous compliance events are obtained based on an instantaneous compliance time window.

[0152] In some specific examples, the instantaneous coincidence time window corresponding to the a priori instantaneous coincidence event is set to 2ns-10ns. When the instantaneous coincidence time window exceeds 10ns, the background signal in the subsequently obtained a priori one-dimensional instantaneous coincidence energy spectrum may be greater than the background signal in the delayed coincidence energy spectrum, resulting in a failure to cancel each other out and causing corresponding errors that interfere with the accuracy of the scattering correction results. Setting the instantaneous coincidence time window within the range of 2ns-10ns helps to eliminate the interference of the background signal in the subsequently obtained a priori one-dimensional instantaneous coincidence energy spectrum.

[0153] In some embodiments, corresponding to step S1111, the delay coincidence time window for the priori delayed coincidence event is set to 2ns-10ns. Specifically, the range of the delayed coincidence time window and the immediate coincidence time window may be the same or different.

[0154] S1124, obtain the prior one-dimensional instantaneous energy spectrum based on the prior instantaneous coincidence event corresponding to the downsampled response line; obtain the prior one-dimensional delayed energy spectrum based on the prior delayed coincidence event corresponding to the downsampled response line.

[0155] For details on how to obtain the prior one-dimensional instantaneous energy spectrum and the prior one-dimensional delayed energy spectrum based on the downsampling response line, please refer to steps S0120 and S1111-S1113 respectively, which will not be repeated here.

[0156] S1125, obtain the difference between the prior one-dimensional instantaneous energy spectrum and the prior one-dimensional delayed energy spectrum, normalize the difference, and obtain the one-dimensional intermediate energy spectrum Y. psf .

[0157] In some embodiments, the difference between the prior one-dimensional instantaneous energy spectrum and the prior one-dimensional delayed energy spectrum is normalized, as detailed in the prior art, which will not be repeated here.

[0158] S1126, based on one-dimensional intermediate energy spectrum Y psf Obtain the fuzzy response matrix p.

[0159] In some embodiments, based on the one-dimensional intermediate energy spectrum Y psf To obtain the fuzzy response matrix p, please refer to existing techniques; details will not be elaborated here.

[0160] Step S120: Based on the correspondence function, the maximum likelihood-expectation-maximization iterative algorithm and the initial iterative value corresponding to the downsampled response line are used to obtain the two one-dimensional gamma photon energy spectra corresponding to each downsampled response line.

[0161] In some embodiments, a one-dimensional gamma photon energy spectrum refers to the energy spectrum composed of the energy of the gamma photons themselves.

[0162] In some embodiments, the maximum likelihood-expectation-maximization (MLEM) iterative algorithm is obtained based on the standard maximum likelihood-expectation-maximization algorithm in PET image reconstruction.

[0163] In some embodiments, step S120 includes:

[0164] S1210, based on the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum, obtains the iterative formula of the maximum likelihood-expectation maximization iterative algorithm.

[0165] In some embodiments, the iterative formula for the maximum likelihood-expectation maximization iterative algorithm is:

[0166]

[0167] Among them, Y i In the diagram, Y represents the one-dimensional probe energy spectrum obtained from scanning the target object corresponding to a single downsampling response line; i represents the i-th vertical bar region (i.e., "bin") in the energy spectrum; b and j represent the b-th and j-th vertical bar regions in the energy spectrum, respectively; k is the iteration number; and p... ij and p ib In this context, p represents the fuzzy response matrix corresponding to the downsampled response line. ij p represents the value in the i-th row and j-th column of the fuzzy response matrix corresponding to this downsampled response line. ib r represents the value in the i-th row and b-th column of the fuzzy response matrix corresponding to this downsampled response line. i This indicates the one-dimensional delayed energy spectrum corresponding to the random coincidence events included in the downsampled response line; The one-dimensional gamma photon energy spectrum corresponding to the j-th vertical stripe region obtained in the (k+1)-th iteration; The first two vertical bars represent the one-dimensional gamma photon energy spectra corresponding to the b-th and j-th vertical regions of the energy spectrum obtained in the k-th iteration, respectively.

[0168] Since the MLEM iterative algorithm does not have a penalty term constraint in its iterative formula, the selection of the initial iteration value in equation (2) is very important.

[0169] In some embodiments, obtaining the initial iteration value includes:

[0170] S1211, obtain the global initial scattered photon energy spectrum.

[0171] In some embodiments, "global" refers to the object obtained using all the coincident response lines corresponding to the probe data. For example, the global initial scattered photon spectrum refers to the initial scattered photon spectrum obtained using all the coincident response lines corresponding to the probe data.

[0172] In some embodiments, step S1211 includes:

[0173] S1211a, Obtain the coincidence response line based on the probe data.

[0174] In some specific embodiments, step S1211a can refer to step S0111, and will not be repeated here.

[0175] S1211b divides all coincidence events corresponding to all coincidence response lines into instant coincidence events and random coincidence events.

[0176] In some embodiments, the specific operation of step S1211b can be referred to the prior art, and will not be repeated here.

[0177] S1211c obtains the global one-dimensional instantaneous energy spectrum based on instantaneous coincidence events; and obtains the global one-dimensional delayed energy spectrum based on delayed coincidence events.

[0178] In some embodiments, the global one-dimensional instantaneous energy spectrum refers to the one-dimensional instantaneous energy spectrum obtained by using all coincident response lines corresponding to the probe data, and the global one-dimensional delayed energy spectrum refers to the one-dimensional delayed energy spectrum obtained by using all coincident response lines corresponding to the probe data.

[0179] In some embodiments, the specific operation of S1211c can be compared with step S1124, and will not be described again here.

[0180] S1211d subtracts the global one-dimensional delayed energy spectrum corresponding to the delayed coincidence event from the global one-dimensional instantaneous energy spectrum corresponding to the instantaneous coincidence event on all response lines to obtain the global one-dimensional undelayed energy spectrum.

[0181] S1211e is the objective function for obtaining the global initial scattered photon energy spectrum based on the global one-dimensional undelayed energy spectrum.

[0182] In some embodiments, the objective function of the global initial scattered photon energy spectrum is:

[0183]

[0184] Among them, A i In this context, A represents the global initial scattered photon energy spectrum; i represents the i-th vertical bar region in the energy spectrum; Ci In this context, C represents the global one-dimensional undelayed energy spectrum, obtained based on the target object, and U... i In this context, U represents the global one-dimensional intermediate energy spectrum Y obtained from the scanning of the entire PET system. psf Global one-dimensional intermediate energy spectrum Y psf Based on the prosthesis acquisition; hel and heu are the lower and upper thresholds of a set high-energy window (specifically set based on experience, usually 511 < hel < heu, where heu is generally 600, 650, or 700, etc.). The tensile coefficient is denoted as .

[0185] In some embodiments, the global one-dimensional intermediate energy spectrum Y psf The acquisition can be referred to in step S1125. Specifically, the one-dimensional intermediate energy spectrum obtained by using all response lines corresponding to the detection data obtained based on the target object is the global one-dimensional intermediate energy spectrum Y. psf .

[0186] It should be understood that C i The high-energy window contains almost entirely unscattered photons, with very few scattered photons. The total number of photons in the high-energy window affects U. i Stretching allows us to estimate the unscattered photon energy spectrum within the complete energy window. Subtracting the estimated unscattered photon energy spectrum (the part after the minus sign in Equation 3 above corresponds to the estimated unscattered photon energy spectrum) from the global one-dimensional undelayed energy spectrum yields the global initial scattered photon energy spectrum.

[0187] It should be understood that the prerequisite for the success of equation (3) is C. i Containing a sufficient number of photons, based on prior information, the number of photons on all coincident response lines is sufficient. The number of photons on a single response line or a single downsampled response line is generally insufficient. If only the number of photons on a single response line or a single downsampled response line is used, S... i It will be affected by statistical noise. If the energy spectrum of a single response line or a single downsampled response line is obtained directly based on equation (3), the obtained S will be significantly affected by statistical noise. i The accuracy is very poor. In the embodiments of this application, due to the global one-dimensional undelayed energy spectrum C i Based on all the corresponding response lines obtained from the probe data, C i The number of photons in the sample is usually sufficient to improve the acquisition of S. i The accuracy.

[0188] In some specific examples, as can be seen from the above, U i The global one-dimensional intermediate energy spectrum of Y obtained by scanning the entire PET system psf U i For obtaining Y, please refer to step S1125. psfThe acquisition of [the resource] will not be elaborated upon here.

[0189] In some embodiments, the instantaneous energy spectrum obtained based on all response lines corresponding to the prior detection data can be called the global prior instantaneous energy spectrum, and the obtained delayed energy spectrum can be called the global prior delayed energy spectrum.

[0190] S1212, deblurring the global initial scattered photon energy spectrum to obtain the global intermediate initial iteration value.

[0191] In some embodiments, the global intermediate initial iteration value refers to the intermediate initial iteration value obtained based on all response lines corresponding to the probe data.

[0192] In some embodiments, step S1212 includes:

[0193] S1212a, based on the maximum likelihood-expectation maximization iterative algorithm, obtains the defuzzing iterative formula.

[0194] In some specific embodiments, the deblurring iterative formula is as follows:

[0195]

[0196] Among them, S i In the diagram, S represents the global one-dimensional probe energy spectrum obtained by scanning the target object. i represents the i-th vertical bar region (i.e., "bin") in the energy spectrum, b and j represent the b-th and j-th vertical bar regions in the energy spectrum, respectively, k is the iteration number, and p... ij and p ib In this context, p represents the fuzzy response matrix corresponding to the downsampled response line. ij Let P be the value in the i-th row and j-th column of the global fuzzy response matrix. ib r represents the value in the i-th row and b-th column of the global fuzzy response matrix. i This represents the global one-dimensional delayed energy spectrum corresponding to all random coincidence events; The global one-dimensional gamma photon energy spectrum corresponding to the j-th vertical stripe region obtained in the (k+1)-th iteration; The global one-dimensional gamma photon energy spectrum is represented by the b-th and j-th vertical regions of the energy spectrum obtained in the k-th iteration, respectively.

[0197]

[0198] Among them, largeconstant represents a very large constant, such as including but not limited to 1,000,000.

[0199] In this embodiment of the application, equation (4) introduces M j This can make S in the formula iIt only includes scattered photons, that is, so that after the equation is blurred, only the scattered part has a value, and the part corresponding to j < 511keV is the scattered part.

[0200] In some embodiments, the global fuzzy response matrix refers to the fuzzy response matrix based on all response lines corresponding to the probe data.

[0201] In some specific examples, the global fuzzy response matrix P is obtained. ij include:

[0202] Prior detection data is obtained based on spurs (such as point sources or line sources);

[0203] All prior coincident response lines are obtained based on prior detection data;

[0204] All prior compliance events corresponding to all prior compliance response lines are divided into global prior immediate compliance events and global prior delayed compliance events.

[0205] Obtain the global prior one-dimensional instantaneous energy spectrum based on global prior instantaneous coincidence events; obtain the global prior one-dimensional delayed energy spectrum based on global prior delayed coincidence events;

[0206] Obtain the difference between the global prior instantaneous energy spectrum and the global prior delayed energy spectrum;

[0207] The difference is normalized to obtain the global prior intermediate energy spectrum Y. psf1 ;

[0208] Based on global prior intermediate energy spectrum Y psf1 Obtain the global fuzzy response matrix P ij .

[0209] In some specific embodiments, how to normalize this difference to obtain the global prior intermediate energy spectrum Y psf1 ; and how to base the global prior intermediate energy spectrum Y psf1 Obtain the global fuzzy response matrix P ij For details, please refer to existing technologies, which will not be elaborated here.

[0210] In some embodiments, the global prior intermediate energy spectrum Y psf1 The acquisition can be referred to in step S1325. Specifically, the one-dimensional intermediate energy spectrum obtained by using all response lines corresponding to the prior detection data obtained based on the spur is the global prior intermediate energy spectrum Y. psf1 .

[0211] S1212b sets the prior initial iteration value and the number of iterations, and iterates through the defuzzification iterative formula until the number of iterations is reached to obtain the scattering part of the global initial energy spectrum.

[0212] In some specific examples, X can be set to... 0 =1, substitute into equation (4) and iterate several times, for example, including but not limited to 50-500 times, to obtain the scattering part X of the global initial energy spectrum. init，sc .

[0213] S1212c estimates the unscattered portion of the global initial energy spectrum based on the scattered portion of the global initial energy spectrum.

[0214] In some embodiments, after obtaining X init，sc In the case of estimating the unscattered portion of the global initial energy spectrum, i.e., to obtain the value of the global initial energy spectrum at j = 511 keV, the unscattered portion of the global initial energy spectrum can be estimated based on an estimation function, which is:

[0215] X j init =δ×Sum(PX) init，sc )×κ,j=511keV (6)

[0216] Among them, X j init The estimated function segment representing the global intermediate initial iterative value function at j = 511 keV, where κ is the ratio of unscattered photons to scattered photons; here, P is the global fuzzy response matrix. ij δ represents the value in the i-th row and j-th column of the global fuzzy response matrix; δ is a hyperparameter.

[0217] In some specific examples, the ratio κ of unscattered photons to scattered photons can be estimated using the following formula:

[0218]

[0219] Here, sum() represents summation, and C and S correspond to C as described above. i S i .

[0220] S1212d obtains the global intermediate initial iteration value based on the scattered part of the global initial energy spectrum and the unscattered part of the global initial energy spectrum.

[0221] In some specific embodiments, since the PET system mainly uses gamma photons, and the energy of gamma photons is 511 keV, when the energy is greater than 511 keV, the corresponding global initial energy spectrum value is 0. Based on this, the global intermediate initial iterative value function is X. init :

[0222]

[0223] Where δ is a hyperparameter, P is the global fuzzy response matrix, and κ is the ratio of unscattered photons to scattered photons.

[0224] In this embodiment of the application, S obtained by equation (3) i There may be errors because equation (3) assumes C i There are no scattered photons in the high-energy window; in fact, C i Scattered photons may still exist in the high-energy window, leading to U i Increased by the elongation factor, S i If the value of κ is reduced, the value obtained based on equation (7) will be overestimated, which can be balanced by the hyperparameter δ. Specifically, the overestimation of κ is balanced by setting δ to a number less than 1. Experiments show that δ between 0.3 and 0.5 achieves good results and basically does not require adjustment. On the other hand, S i The initial global scattered photon energy spectrum is the energy spectrum obtained based on all response lines detected by the target object. Response lines that do not pass through the target object are considered to be all scattering events. Response lines that pass through the target object contain both scattering events and true events, i.e., unscattered events. Therefore, the proportion of scattering events that pass through the response lines of the target object is lower than that of scattering events on all response lines. The unscattered part can also be obtained more accurately through the hyperparameter δ balancing.

[0225] S1213, stretch the global intermediate initial iteration value to obtain the initial iteration value corresponding to each downsampled response line.

[0226] In some embodiments, step S1213 includes:

[0227] The global intermediate initial iteration value is stretched using a stretching function to obtain the initial iteration value corresponding to each downsampled response line.

[0228] In some specific examples, the stretching function is:

[0229]

[0230] Among them, X 0 The initial iteration value is given by `sum()`, where `P` represents the summation function, `X` is the global fuzzy response matrix, and `X` is the initial iteration value. init is the global intermediate initial iteration value; Y is the one-dimensional probe energy spectrum obtained based on the target object scan corresponding to a single downsampled response line; r represents the one-dimensional delayed energy spectrum corresponding to the random coincidence events included in a single response line.

[0231] In this embodiment, by stretching, an initial iteration value X matching the number of single events included in the downsampling response line can be obtained. 0 .

[0232] S1220, Set the number of iterations, and iterate using the maximum likelihood-expectation maximization iterative algorithm formula based on the obtained initial iteration value until the set number of iterations is reached, and obtain the one-dimensional gamma photon energy spectrum corresponding to the one-dimensional detector energy spectrum.

[0233] In some specific embodiments, the number of iterations is set according to needs, such as 20-60 iterations. A good one-dimensional gamma photon spectrum is usually obtained within 60 iterations.

[0234] Step S130: Obtain the number of unscattered single events corresponding to each downsampled response line based on the one-dimensional gamma photon energy spectrum.

[0235] In some embodiments, the count value corresponding to the energy of the gamma photons that have not been scattered in the one-dimensional gamma photon energy spectrum is used as the number of unscattered single events; the sum of the number of unscattered single events in the two one-dimensional gamma photon energy spectra corresponding to each downsampled response line is used as the number of unscattered events corresponding to the corresponding downsampled response line.

[0236] Step S140: Calculate the number of scattering coincidence events corresponding to each downsampled response line based on the total number of single events and the number of unscattered single events corresponding to each downsampled response line.

[0237] In some embodiments, step S160 is performed using the following formula:

[0238] SC = S tot -S unsc

[0239] or

[0240] or

[0241] Among them, S tot The total number of single events is twice the total number of coincident events corresponding to all response lines; S unsc SC represents the number of unscattered single events; SC represents the number of scattering coincidence events.

[0242] In some embodiments, step S140 includes:

[0243] Count the total number of single events and the number of unscattered single events corresponding to each downsampling response line.

[0244] In some specific examples, the total number of single events corresponding to each downsampling response line is counted, including:

[0245] The total number of single events corresponding to each downsampled response line is calculated by counting all response lines included in each downsampled response line and then multiplying the total number of coincident events corresponding to all response lines by twice the total number of coincident events.

[0246] Step S150: Perform upsampling processing on the downsampled response line to obtain the upsampled response line.

[0247] In some embodiments, upsampling the downsampled response line includes:

[0248] The downsampled response lines are processed using bilinear interpolation, 4D linear interpolation, or 5D linear interpolation to obtain the corresponding upsampled response lines. Specific procedures can be found in existing techniques and will not be elaborated upon here.

[0249] Step S160: Calculate the number of scattering coincidence events corresponding to each upsampled response line.

[0250] In some embodiments, the scintillation crystals included in the detection module of the PET system are referred to as upsampling crystals, and the set of scintillation crystals corresponding to the downsampling response lines obtained by downsampling processing is referred to as downsampling crystals, for example, such as Figure 2 As shown, the downsampling crystal to which the upsampling crystal (represented by the small square within the dashed box) belongs is denoted as "downsampling center crystal c"; the downsampling crystal closest to the upsampling crystal along its axis is denoted as "downsampling axial adjacent crystal a"; the downsampling crystal closest to the upsampling crystal in its radial direction is denoted as "downsampling radial adjacent crystal t"; and the downsampling crystal closest to the upsampling crystal in the diagonal direction is denoted as "downsampling relative crystal ta". For example... Figure 2 As shown, let the axial length of the downsampling crystal be D. a The radial length is D t An upsampling response line has two upsampling crystals, distinguished by the superscripts "i" and "j". Taking the scintillation crystal labeled "i" as an example, let the radial and axial distances from the center of this upsampling crystal to the center of the "center crystal c" be respectively... The weights corresponding to the four downsampling crystals are as follows:

[0251] Downsampling center crystal c”:

[0252] Downsampling axial adjacent crystal a”:

[0253] Downsampling radial adjacent crystal t”:

[0254] Downsampling relative to crystal ta”: Similarly, for the four downsampling crystals corresponding to the upsampling crystal with the superscript "j", we also have:

[0255]

[0256]

[0257]

[0258]

[0259] In some specific embodiments, when processing the downsampled response line using the 4D linear interpolation method, the following formula is used to calculate the number of scattering coincidence events corresponding to the upsampled response line:

[0260]

[0261] in, The number of scattering coincidence events characterizing the upsampled response line composed of upsampled crystals i and j after 4D linear interpolation; S i This represents the set of downsampling center crystal, downsampling axial adjacent crystal, downsampling radial adjacent crystal, and downsampling relative crystal corresponding to upsampling crystal i; K and L represent the crystal numbers corresponding to the downsampling response lines; Characterize the number of scattering coincidence events on the downsampled response line composed of downsampled crystals K and L; This represents the weight of the downsampling crystal K relative to the upsampling crystal i; W represents the weight of the downsampling crystal L relative to the upsampling crystal j; tot The normalization factor is characterized as follows:

[0262]

[0263] Where N is the number of upsampled response lines contained in a downsampled response line.

[0264] In some other embodiments, step S180 includes:

[0265] Calculate the ratio of the total number of scattering coincidence events to the total number of coincidence events on each downsampled response line;

[0266] The number of scattering coincidence events corresponding to each upsampled response line is calculated based on the number of coincidence events on each upsampled response line contained in each downsampled response line and the aforementioned proportion.

[0267] In this embodiment, the MLEM algorithm is used for scattering correction. This method does not require the hyperparameter δ, thus eliminating the need to set the hyperparameter δ within a defined range. Compared with the existing one-step lag iterative algorithm OSL-EDR (One Step Latereliminate detector response) and the optimization transfer eliminate detector response (Optimization transfer eliminate detector response) algorithm OT-EDR, both OSL-EDR and OT-EDR require the selection of an appropriate hyperparameter δ to achieve better results. The optimal β requires some preliminary experiments for adjustment, and different spurs of different sizes require different δ values. In this embodiment, the MLEM algorithm is used for scattering correction, which is minimally affected by the hyperparameter δ and requires almost no adjustment, making scattering correction more convenient.

[0268] As can be seen from the above, some embodiments of this application consider various scattering scenarios, such as random scattering and external scattering events that do not penetrate the target object, resulting in higher accuracy of the scattering correction results and thus improving the quality of the subsequent reconstructed image. Furthermore, compared with existing double scattering simulation methods and Monte Carlo simulation methods, the scattering correction method according to this application has the advantages of simpler calculation process and lower computational load.

[0269] Figure 3 This is an exemplary flowchart of a scattering correction method 200 according to some embodiments of this application.

[0270] See also Figure 3 The scattering correction method 200 may include the following steps:

[0271] S210, the function that obtains the correspondence between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum;

[0272] S220, based on the correspondence function, uses the maximum likelihood-expectation-maximization iterative algorithm and the initial iterative value corresponding to the response line to obtain the two one-dimensional gamma photon energy spectra corresponding to each response line;

[0273] S230, based on the one-dimensional gamma photon energy spectrum, obtains the number of unscattered single events corresponding to each response line;

[0274] S240, based on the total number of single events corresponding to each response line and the number of unscattered single events, calculate the number of scattering coincidence events corresponding to each response line.

[0275] In some embodiments, prior to step S210, the following step is further included:

[0276] S0210, Obtain the response line based on the probe data;

[0277] S0220, two one-dimensional probe energy spectra are obtained based on each response line.

[0278] Unlike the embodiment of scattering correction method 100, which obtains response lines based on probe data, downsamples the response lines to obtain downsampled response lines, processes the downsampled response lines using a maximum likelihood-expectation-maximization iterative algorithm to obtain the number of scattering coincidence events corresponding to the downsampled response lines, and then upsamples the downsampled response lines to obtain the number of scattering coincidence events for each response line, this embodiment directly processes the response lines obtained based on probe data using a maximum likelihood-expectation-maximization iterative algorithm to obtain the number of scattering coincidence events for each response line.

[0279] In the embodiments of this application, the scattering correction method 200 may selectively combine features of the scattering correction method 100 or other methods, or vice versa.

[0280] Figure 4 This is an exemplary flowchart of an image reconstruction method according to some embodiments of this application.

[0281] See also Figure 4 The image reconstruction method 300 may include the following steps:

[0282] S310, the scattering correction method described in the above embodiments is used to obtain the number of scattering coincidence events corresponding to each upsampled response line;

[0283] S320 corrects the scattering events based on the number of scattering coincidence events to obtain a reconstructed image.

[0284] In some specific embodiments, an iterative image reconstruction algorithm can be used to reconstruct the image based on the number of scattering coincidence events, thereby obtaining the reconstructed image.

[0285] Step S320 can be found in existing technology and will not be described in detail here.

[0286] Figure 5 This is an exemplary block diagram of a scattering correction device according to some embodiments of this application. For example... Figure 5 As shown, the scattering correction device 400 may include:

[0287] The correspondence acquisition module 410 is configured to acquire the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum;

[0288] The gamma photon spectrum acquisition module 420 is configured to acquire two one-dimensional gamma photon spectra corresponding to each downsampled response line based on the correspondence function, using the maximum likelihood-expectation-maximization iterative algorithm and the initial iterative value corresponding to the downsampled response line.

[0289] The module 430 for obtaining the number of unscattered single events is configured to obtain the number of unscattered single events corresponding to each downsampled response line based on a one-dimensional gamma photon energy spectrum.

[0290] The scattering coincidence event acquisition module 440 is configured to calculate the scattering coincidence event number corresponding to each downsampled response line based on the total number of single events corresponding to each downsampled response line and the number of unscattered single events.

[0291] The upsampling module 450 is configured to perform upsampling processing on the downsampling response line to obtain the upsampling response line;

[0292] The scattering coincidence event count calculation module 460 is configured to calculate the number of scattering coincidence events corresponding to each upsampled response line.

[0293] In some embodiments, the correspondence acquisition module 410 includes a delayed energy spectrum acquisition module, which is configured to acquire a one-dimensional delayed energy spectrum; and a fuzzy response matrix acquisition module, which is configured to acquire prior detection data based on a spur, and acquire a global fuzzy response matrix and the fuzzy response matrix corresponding to each downsampled response line based on the prior detection data.

[0294] In some embodiments, the gamma photon energy spectrum acquisition module 420 includes a maximum likelihood-expectation-maximization iterative formula acquisition module, configured to acquire the maximum likelihood-expectation-maximization iterative formula based on the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum; an initial iteration value acquisition module, configured to acquire an initial iteration value; and an iteration module, configured to set an iteration number and iterate based on the acquired initial iteration value using the maximum likelihood-expectation-maximization iterative formula until the set iteration number is reached, thereby acquiring the one-dimensional gamma photon energy spectrum corresponding to the one-dimensional detector energy spectrum.

[0295] In some specific embodiments, the initial iteration value acquisition module includes a global initial scattered photon energy spectrum, which is configured to acquire the global initial scattered photon energy spectrum; an intermediate initial iteration value acquisition module, which is configured to perform deblurring processing on the global initial scattered photon energy spectrum to acquire global intermediate initial iteration values; and a stretching module, which stretches the global intermediate initial iteration values ​​to acquire the initial iteration values ​​corresponding to each downsampled response line.

[0296] In some embodiments, the scattering correction device 400 further includes a statistics module configured to count the total number of single events and the number of unscattered single events corresponding to each downsampling response line.

[0297] In the embodiments of this application, the scattering correction device 400 may selectively incorporate features of the scattering correction method 100 or 200 or other methods, or vice versa.

[0298] In the embodiments of this application, the scattering correction device 400 can be used to implement the scattering correction method 100 or the methods described in other embodiments herein, and can be selectively combined with features of the scattering correction method 200 or other methods, or vice versa.

[0299] Figure 6 This is an exemplary block diagram of a scattering correction device according to other embodiments of this application. For example... Figure 6 As shown, the scattering correction device 500 may include:

[0300] The correspondence acquisition module 510 is configured to acquire the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum;

[0301] The gamma photon energy spectrum acquisition module 520 is configured to acquire two one-dimensional gamma photon energy spectra corresponding to each response line based on the correspondence function, using the maximum likelihood-expectation-maximization iterative algorithm and the initial iterative values ​​corresponding to the response lines.

[0302] The module 530 for obtaining the number of unscattered single events is configured to obtain the number of unscattered single events corresponding to each response line based on a one-dimensional gamma photon energy spectrum.

[0303] The scattering coincidence event acquisition module 540 is configured to calculate the scattering coincidence event number corresponding to each response line based on the total number of single events corresponding to each response line and the number of unscattered single events.

[0304] Unlike the scattering correction device 400 embodiment, which obtains response lines based on probe data, downsamples the response lines to obtain downsampled response lines, processes the downsampled response lines using a maximum likelihood-expectation-maximization iterative algorithm to obtain the number of scattering coincidence events corresponding to the downsampled response lines, and then upsamples the downsampled response lines to obtain the number of scattering coincidence events for each response line, this embodiment directly processes the response lines obtained based on probe data using a maximum likelihood-expectation-maximization iterative algorithm to obtain the number of scattering coincidence events for each response line.

[0305] In the embodiments of this application, the scattering correction device 500 can be used to implement the scattering correction method 200 or the method described in other embodiments herein, and can be selectively combined with the features of the scattering correction device 400 or other devices, or selectively combined with the features of the scattering correction method 100 or 200 or other methods, and vice versa.

[0306] Figure 7 These are exemplary block diagrams of an image reconstruction apparatus according to some embodiments of this application. Figure 7 As shown, the image reconstruction apparatus 600 may include:

[0307] The scattering coincidence event acquisition module 610 is configured to acquire the number of scattering coincidence events corresponding to each upsampled response line based on the scattering correction method described in any of the above embodiments.

[0308] The reconstruction module 620 is configured to correct scattering events based on the number of scattering coincidence events and obtain a reconstructed image.

[0309] In the embodiments of this application, the image reconstruction apparatus 600 can be used to implement the image reconstruction method 300 or the methods described in other embodiments herein, and can selectively combine features of the image reconstruction method 300 or other methods, or vice versa.

[0310] In some embodiments of this application, the image reconstruction apparatus 600 may also include components or features of the scattering correction apparatus 400, 500 in a manner that is not contradictory, and vice versa.

[0311] In some embodiments, this application also provides a digitizing device, which includes the apparatus described in any one of the above embodiments.

[0312] In some embodiments, this application also provides a digital device that may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, can implement the steps of the method described in any of the above embodiments.

[0313] Although not shown, some embodiments also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in any of the above embodiments. The computer program includes various program modules / units constituting the apparatus according to embodiments of this application, and when executed, the computer program comprised of these program modules / units is capable of performing functions corresponding to the various steps in the methods described in the above embodiments. The computer program can also run on electronic devices as described in embodiments of this application.

[0314] The basic concepts have been described herein. It is obvious that the detailed disclosure above is merely illustrative and does not constitute a limitation of this application. Although not explicitly stated herein, various modifications, improvements, and corrections may be made to this application by those skilled in the art. Such modifications, improvements, and corrections are suggested in this application and therefore remain within the spirit and scope of the exemplary embodiments of this application.

[0315] Furthermore, this application uses specific terms to describe its embodiments. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic related to at least one embodiment of this application. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this application do not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of this application can be appropriately combined.

[0316] Furthermore, those skilled in the art will understand that aspects of this application can be described and illustrated through several patentable types or situations, including any new and useful combination of processes, machines, products, or substances, or any new and useful improvements thereof. Accordingly, aspects of this application can be implemented entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. All of the above hardware or software may be referred to as a “data block,” “module,” “engine,” “unit,” “component,” or “system.” Furthermore, aspects of this application may manifest as a computer product located on one or more computer-readable media, the product including computer-readable program code.

[0317] Computer storage media may contain a propagated data signal containing computer program code, for example, on baseband or as part of a carrier wave. This propagated signal may take various forms, including electromagnetic, optical, and suitable combinations thereof. Computer storage media can be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, apparatus, or device to enable communication, propagation, or transmission of a program for use. The program code located on the computer storage medium can be propagated through any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0318] The computer program code required for the operation of each part of this application can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby, and Groovy, or other programming languages. This program code can run entirely on the user's computer, or as a standalone software package on the user's computer, or partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any network, such as a local area network (LAN) or wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as Software as a Service (SaaS).

[0319] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this application are not intended to limit the order of the processes and methods of this application. Although the foregoing disclosure has discussed some currently considered useful embodiments of the invention through various examples, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the substance and scope of the embodiments of this application. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely through software solutions, such as installing the described system on existing servers or mobile devices.

[0320] Similarly, it should be noted that, in order to simplify the description of the present application and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of the embodiments of the present application sometimes combines multiple features into a single embodiment, drawing, or description thereof. However, this disclosure method does not imply that the subject matter of the application requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of the single embodiments disclosed above.

[0321] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of scope in some embodiments of this application are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0322] For each patent, patent application, patent application publication, and other material such as articles, books, specifications, publications, and documents referenced in this application, the entire contents of that patent are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this application, as well as documents that limit the broadest scope of the claims in this application (currently or subsequently appended to this application). It should be noted that if there are any inconsistencies or conflicts between the descriptions, definitions, and / or terminology used in the supplementary materials of this application and the content of this application, the descriptions, definitions, and / or terminology used in this application shall prevail.

[0323] Finally, it should be noted that the above descriptions are merely exemplary embodiments of this application and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A scattering correction method, characterized in that, The scattering correction method includes: Obtain the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum; Based on the correspondence function, the maximum likelihood-expectation-maximization iterative algorithm and the initial iterative value corresponding to the downsampled response line are used to obtain the two one-dimensional gamma photon energy spectra corresponding to each downsampled response line. The number of unscattered single events corresponding to each downsampled response line is obtained based on the one-dimensional gamma photon energy spectrum. Based on the total number of single events and the number of unscattered single events corresponding to each downsampled response line, calculate the number of scattering coincidence events corresponding to each downsampled response line. The downsampled response line is upsampled to obtain the upsampled response line; Calculate the number of scattering coincidence events corresponding to each upsampled response line; Determining the initial iteration value includes: obtaining the global initial scattered photon energy spectrum; performing deblurring on the global initial scattered photon energy spectrum to obtain the global intermediate initial iteration value; and stretching the global intermediate initial iteration value to obtain the initial iteration value corresponding to each downsampled response line.

2. The scattering correction method according to claim 1, characterized in that, Before obtaining the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum, the following steps are also included: Obtain the downsampling response line based on the probe data; Two one-dimensional probe energy spectra are obtained based on each downsampling response line.

3. The scattering correction method according to claim 2, characterized in that, The downsampling response line is acquired based on the probe data and includes: Obtain the response line based on the probe data; The coincident response line is downsampled to obtain a downsampled response line; The detection data is obtained based on the target object.

4. The scattering correction method according to claim 2, characterized in that, The one-dimensional probe energy spectrum is obtained based on each downsampled response line, including: Break down all coincidence events corresponding to the downsampling response line into single events; Based on the location information of a single event, the single event is divided into two parts, and each part corresponds to a one-dimensional detection energy spectrum. Based on the two parts of single events, two one-dimensional probe energy spectra are obtained.

5. The scattering correction method according to claim 4, characterized in that, The location information of a single event includes the location information of the downsampling detection module that detected the single event.

6. The scattering correction method according to claim 1, characterized in that, The correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum is as follows: ， in, Characterizing the fuzzy response matrix, This represents the expected value of the one-dimensional probe energy spectrum; Represents the one-dimensional gamma photon energy spectrum. This represents the one-dimensional delayed energy spectrum corresponding to a random coincidence event.

7. The scattering correction method according to claim 6, characterized in that, The one-dimensional delayed energy spectrum corresponding to the random coincidence event is obtained through delayed coincidence event estimation.

8. The scattering correction method according to claim 7, characterized in that, Obtaining a one-dimensional delayed energy spectrum includes: All delayed coincidence events in the downsampled response line are obtained based on the delayed coincidence window; the delayed coincidence events are then broken down into delayed single events. Based on the location information of the delayed single event, the delayed single event is divided into two parts, each part corresponding to a one-dimensional delayed energy spectrum; Based on the two-part delayed single event, two one-dimensional delayed energy spectra are obtained; The location information of the delayed single event includes the location information of the downsampling detection module that detected the delayed single event.

9. The scattering correction method according to claim 6, characterized in that, Obtaining the fuzzy response matrix include: Acquiring prior detection data based on prostheses; Obtain the a priori downsampling response line based on prior detection data; All prior compliance events corresponding to the prior downsampling response line are divided into prior immediate compliance events and prior delayed compliance events. A prior one-dimensional instantaneous energy spectrum is obtained based on the prior instantaneous coincidence event corresponding to the downsampled response line; a prior one-dimensional delayed energy spectrum is obtained based on the prior delayed coincidence event corresponding to the downsampled response line. Obtain the difference between the prior one-dimensional instantaneous energy spectrum and the prior one-dimensional delayed energy spectrum, and normalize the difference to obtain the one-dimensional intermediate energy spectrum. The fuzzy response matrix is ​​obtained based on the one-dimensional intermediate energy spectrum.

10. The scattering correction method according to claim 1, characterized in that, Obtain the two one-dimensional gamma photon energy spectra corresponding to each downsampled response line, including: Based on the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum, the iterative formula of the maximum likelihood-expectation maximization iterative algorithm is obtained. The number of iterations is set, and the algorithm iterates using the maximum likelihood-expectation-maximization algorithm based on the obtained initial iteration value until the set number of iterations is reached, thereby obtaining the one-dimensional gamma photon energy spectrum corresponding to the one-dimensional detector energy spectrum.

11. The scattering correction method according to claim 10, characterized in that, The iterative formula for the maximum likelihood-expectation maximization iterative algorithm is: ， in, In Characterizes the one-dimensional probe energy spectrum obtained from scanning the target object, corresponding to a single downsampling response line. Characterizing the first in the energy spectrum A vertical strip area, , Characterizing the first in the energy spectrum respectively The, the A vertical strip area, It is the number of iterations. and In This is the fuzzy response matrix corresponding to the downsampled response line. The fuzzy response matrix corresponding to this downsampled response line is the first... line, number The value of the column, The fuzzy response matrix corresponding to this downsampled response line is the first... line, number The value of the column; This indicates the one-dimensional delayed energy spectrum corresponding to the random coincidence events included in the downsampled response line; For the first The energy spectrum obtained in the +1 iteration is the first... The one-dimensional gamma photon energy spectrum corresponding to each vertical bar region; , Respectively characterize the first The energy spectrum is obtained in the nth iteration. The, the The vertical bar regions correspond to the one-dimensional gamma photon energy spectrum.

12. The scattering correction method according to claim 1, characterized in that, Obtain the global initial scattered photon energy spectrum, including: Obtain the response line based on the probe data; All matching events corresponding to all matching response lines are divided into immediate matching events and random matching events; Obtain the global one-dimensional instantaneous energy spectrum based on instantaneous coincidence events; obtain the global one-dimensional delayed energy spectrum based on delayed coincidence events; Subtract the global one-dimensional delayed energy spectrum corresponding to the delayed coincidence event from the global one-dimensional instant energy spectrum corresponding to the instant coincidence event of all responses online to obtain the global one-dimensional undelayed energy spectrum; The objective function for obtaining the global initial scattered photon energy spectrum is based on the global one-dimensional undelayed energy spectrum.

13. The scattering correction method according to claim 12, characterized in that, The objective function for the global initial scattered photon energy spectrum is: ， in, In The initial global scattered photon energy spectrum; Characterizing the first in the energy spectrum A vertical strip area; In The global one-dimensional undelayed energy spectrum is obtained based on the target object. In The global one-dimensional intermediate energy spectrum was obtained from the entire PET system scan; the global one-dimensional intermediate energy spectrum was obtained based on the prosthesis. and The lower and upper thresholds of the high-energy window are set. The tensile coefficient is denoted as .

14. The scattering correction method according to claim 13, characterized in that, Get include: Acquiring prior detection data based on prostheses; All prior coincident response lines are obtained based on prior detection data; All prior compliance events corresponding to all prior compliance response lines are divided into global prior immediate compliance events and global prior delayed compliance events. Obtain the global prior one-dimensional instantaneous energy spectrum based on global prior instantaneous coincidence events; obtain the global prior one-dimensional delayed energy spectrum based on global prior delayed coincidence events; Obtain the difference between the global prior instantaneous energy spectrum and the global prior delayed energy spectrum. .

15. The scattering correction method according to claim 1, characterized in that, The global initial scattered photon energy spectrum is deblurred to obtain global intermediate initial iteration values, including: The defuzzing iterative formula is obtained based on the maximum likelihood-expectation maximization iterative algorithm; Set the prior initial iteration value and the number of iterations, and iterate through the defuzzification iterative formula until the number of iterations is reached to obtain the scattering part of the global initial energy spectrum; Estimate the unscattered portion of the global initial energy spectrum based on the scattered portion of the global initial energy spectrum; Based on the scattered portion and the unscattered portion of the global initial energy spectrum, the global intermediate initial iteration value is obtained.

16. The scattering correction method according to claim 15, characterized in that, The iterative formula for deblurring is: ， in, In The characterization is based on the global one-dimensional probe energy spectrum obtained by scanning the target object. Characterizing the first in the energy spectrum A vertical strip area, , Characterizing the first in the energy spectrum respectively The, the A vertical strip area, It is the number of iterations. and In This is the fuzzy response matrix corresponding to the downsampled response line. The first fuzzy response matrix is ​​the global fuzzy response matrix. line, number The value of the column, The first fuzzy response matrix is ​​the global fuzzy response matrix. line, number The value of the column; This represents the global one-dimensional delayed energy spectrum corresponding to all random coincidence events; For the first The +1st iteration obtained the first The global one-dimensional gamma photon energy spectrum corresponding to each vertical bar region; , Representing the k-th iteration, respectively... The, the The global one-dimensional gamma photon energy spectrum corresponding to each vertical bar region; ， in, Characterization constant.

17. The scattering correction method according to claim 16, characterized in that, Obtain the global fuzzy response matrix include: Acquiring prior detection data based on prostheses; All prior coincident response lines are obtained based on prior detection data; All prior compliance events corresponding to all prior compliance response lines are divided into global prior immediate compliance events and global prior delayed compliance events. Obtain the global prior one-dimensional instantaneous energy spectrum based on global prior instantaneous coincidence events; obtain the global prior one-dimensional delayed energy spectrum based on global prior delayed coincidence events; Obtain the difference between the global prior instantaneous energy spectrum and the global prior delayed energy spectrum; The difference is normalized to obtain the global prior intermediate energy spectrum; The global fuzzy response matrix is ​​obtained based on the global prior intermediate energy spectrum.

18. The scattering correction method according to claim 15, characterized in that, The unscattered portion of the global initial energy spectrum is estimated using an estimation function, which is: ， in, This is the global fuzzy response matrix; For hyperparameters; This represents the ratio of unscattered photons to scattered photons. in, To express summation, and Corresponding to , .

19. The scattering correction method according to claim 15, characterized in that, The representation function of the global intermediate initial iteration value is: ， in, It's a hyperparameter. The global fuzzy response matrix, This represents the ratio of unscattered photons to scattered photons.

20. The scattering correction method according to claim 1, characterized in that, Stretch the intermediate initial iteration values ​​to obtain the initial iteration values ​​corresponding to each downsampled response line, including: The global intermediate initial iteration value is stretched using a stretching function to obtain the initial iteration value corresponding to each downsampled response line. The stretching function is: ， in, The initial iteration value, To express summation, The global fuzzy response matrix, This is the global intermediate initial iteration value; This represents the one-dimensional probe energy spectrum obtained from target object scanning, corresponding to a single downsampling response line. This represents the one-dimensional delayed energy spectrum corresponding to the random coincidence events included in a single response line.

21. The scattering correction method according to claim 1, characterized in that, The number of unscattered events corresponding to each downsampled response line is obtained based on the one-dimensional gamma photon energy spectrum, including: The count value corresponding to the energy of the gamma photons that did not undergo scattering in the one-dimensional gamma photon energy spectrum is taken as the number of unscattered single events; the sum of the number of unscattered single events in the two one-dimensional gamma photon energy spectra corresponding to each downsampled response line is taken as the number of unscattered events corresponding to the corresponding downsampled response line.

22. The scattering correction method according to claim 1, characterized in that, The number of scattering coincidence events corresponding to each downsampled response line is calculated using the following formula: ， or , or ; in, The total number of single events is twice the total number of events corresponding to all response lines, which is the total number of single events corresponding to the downsampled response line. SC represents the number of unscattered single events; SC represents the number of scattering coincidence events.

23. The scattering correction method according to claim 1, characterized in that, Upsampling is performed on the downsampled response line, including: The downsampled response lines are processed using bilinear interpolation, 4D linear interpolation, or 5D linear interpolation to obtain the upsampled response lines corresponding to each downsampled response line.

24. The scattering correction method according to claim 23, characterized in that, When processing the downsampled response line using the 4D linear interpolation method, the number of scattering coincidence events corresponding to the upsampled response line is calculated using the following formula: ； in, Characterized by 4D linear interpolation by upsampling crystal and The number of scattering coincidence events that constitute the upsampled response line; Indicates upsampling crystal The set of the corresponding downsampling center crystal, downsampling axial adjacent crystal, downsampling radial adjacent crystal, and downsampling relative crystal; and This indicates the crystal number corresponding to the downsampled response line; Characterization of downsampling crystal and The number of scattering coincidence events on the downsampled response line; This indicates that for upsampling crystals Regarding downsampling crystals The weights; This indicates that for upsampling crystals Regarding downsampling crystals The weights; The normalization factor is characterized as follows: ； in The number of upsampled response lines contained in a downsampled response line.

25. The scattering correction method according to claim 1, characterized in that, Calculate the number of scattering coincidence events corresponding to each upsampled response line, including: Calculate the ratio of the total number of scattering coincidence events to the total number of coincidence events on each of the downsampled response lines; The number of scattering coincidence events corresponding to each upsampled response line is calculated based on the number of coincidence events on each upsampled response line contained in each downsampled response line and the proportion.

26. A scattering correction method, characterized in that, The scattering correction method includes: Obtain the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum; Based on the correspondence function, the maximum likelihood-expectation-maximization iterative algorithm and the initial iterative values ​​corresponding to the response lines are used to obtain the two one-dimensional gamma photon energy spectra corresponding to each response line. The number of unscattered single events corresponding to each response line is obtained based on the one-dimensional gamma photon energy spectrum. Based on the total number of single events corresponding to each response line and the number of unscattered single events, calculate the number of scattering coincidence events corresponding to each response line. Determining the initial iteration value includes: obtaining the global initial scattered photon energy spectrum; performing deblurring on the global initial scattered photon energy spectrum to obtain the global intermediate initial iteration value; and stretching the global intermediate initial iteration value to obtain the initial iteration value corresponding to each response line.

27. The scattering correction method according to claim 26, characterized in that, The one-dimensional gamma photon energy spectrum is obtained based on each response line, including: Based on the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum, the iterative formula of the maximum likelihood-expectation maximization iterative algorithm is obtained. The number of iterations is set, and the algorithm iterates using the maximum likelihood-expectation-maximization algorithm based on the obtained initial iteration value until the set number of iterations is reached, thereby obtaining the one-dimensional gamma photon energy spectrum corresponding to the one-dimensional detector energy spectrum.

28. The scattering correction method according to claim 26, characterized in that, Obtain the global initial scattered photon energy spectrum, including: Obtain the response line based on the probe data; All matching events corresponding to all matching response lines are divided into immediate matching events and random matching events; Obtain the global one-dimensional instantaneous energy spectrum based on instantaneous coincidence events; obtain the global one-dimensional delayed energy spectrum based on delayed coincidence events; Subtract the global one-dimensional delayed energy spectrum corresponding to the delayed coincidence event from the global one-dimensional instant energy spectrum corresponding to the instant coincidence event of all responses online to obtain the global one-dimensional undelayed energy spectrum; The objective function for obtaining the global initial scattered photon energy spectrum is based on the global one-dimensional undelayed energy spectrum.

29. The scattering correction method according to claim 26, characterized in that, The global initial scattered photon energy spectrum is deblurred to obtain global intermediate initial iteration values, including: The defuzzing iterative formula is obtained based on the maximum likelihood-expectation maximization iterative algorithm; Set the prior initial iteration value and the number of iterations, and iterate through the defuzzification iterative formula until the number of iterations is reached to obtain the scattering part of the global initial energy spectrum; Estimate the unscattered portion of the global initial energy spectrum based on the scattered portion of the global initial energy spectrum; Based on the scattered portion and the unscattered portion of the global initial energy spectrum, the global intermediate initial iteration value is obtained.

30. The scattering correction method according to claim 26, characterized in that, The number of unscattered events corresponding to each response line is obtained based on the one-dimensional gamma photon energy spectrum, including: The count value corresponding to the energy of the gamma photons that did not undergo scattering in the one-dimensional gamma photon energy spectrum is taken as the number of unscattered single events; the sum of the number of unscattered single events in the two one-dimensional gamma photon energy spectra corresponding to each response line is taken as the number of unscattered events corresponding to the corresponding response line.

31. The scattering correction method according to claim 26, characterized in that, The number of scattering coincidence events corresponding to each response line is calculated using the following formula: ， or , or ; in, The total number of single events is twice the total number of events corresponding to all response lines, which is the total number of single events corresponding to the downsampled response line. SC represents the number of unscattered single events; SC represents the number of scattering coincidence events.

32. The scattering correction method according to claim 26, characterized in that, Before obtaining the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum, the following steps are also included: Response lines are obtained based on probe data; Two one-dimensional probe energy spectra are obtained based on each response line.

33. An image reconstruction method, characterized in that, The image reconstruction method includes: The number of scattering coincidence events corresponding to each response line is obtained based on the scattering correction method according to any one of claims 1-32; Based on the number of scattering coincidence events, an iterative image reconstruction algorithm is used to reconstruct the image and obtain the reconstructed image.

34. A scattering correction device, characterized in that, The scattering correction device includes: The correspondence acquisition module is configured to acquire the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum; The gamma photon spectrum acquisition module is configured to acquire two one-dimensional gamma photon spectra corresponding to each downsampled response line based on a correspondence function, using a maximum likelihood-expectation-maximization iterative algorithm and initial iterative values ​​corresponding to the downsampled response lines. Acquiring the initial iterative values ​​includes: acquiring the global initial scattered photon spectrum; deblurring the global initial scattered photon spectrum to acquire global intermediate initial iterative values; and stretching the global intermediate initial iterative values ​​to acquire the initial iterative values ​​corresponding to each downsampled response line. The module for obtaining the number of unscattered single events is configured to obtain the number of unscattered single events corresponding to each downsampled response line based on a one-dimensional gamma photon energy spectrum. The module for obtaining the number of scattering coincidence events is configured to calculate the number of scattering coincidence events corresponding to each downsampled response line based on the total number of single events corresponding to each downsampled response line and the number of unscattered single events. The upsampling module is configured to perform upsampling processing on the downsampling response line to obtain the upsampling response line; The scattering coincidence event count calculation module is configured to calculate the number of scattering coincidence events corresponding to each upsampled response line.

35. A scattering correction device, characterized in that, The scattering correction device includes: The correspondence acquisition module is configured to acquire the correspondence function between the one-dimensional detector energy spectrum and the one-dimensional gamma photon energy spectrum; The gamma photon spectrum acquisition module is configured to acquire two one-dimensional gamma photon spectra corresponding to each response line based on the correspondence function, using a maximum likelihood-expectation-maximization iterative algorithm and initial iteration values ​​corresponding to the response lines. Acquiring the initial iteration values ​​includes: acquiring the global initial scattered photon spectrum; deblurring the global initial scattered photon spectrum to acquire global intermediate initial iteration values; and stretching the global intermediate initial iteration values ​​to acquire the initial iteration values ​​corresponding to each downsampled response line. The module for obtaining the number of unscattered single events is configured to obtain the number of unscattered single events corresponding to each response line based on a one-dimensional gamma photon energy spectrum. The module for obtaining the number of scattering coincidence events is configured to calculate the number of scattering coincidence events corresponding to each response line based on the total number of single events corresponding to each response line and the number of unscattered single events.

36. An image reconstruction apparatus, characterized in that, The image reconstruction apparatus includes: A scattering coincidence event acquisition module is configured to acquire the number of scattering coincidence events corresponding to each response line based on the scattering correction device described in claim 34 or 35. The reconstruction module is configured to use an iterative image reconstruction algorithm based on the number of scattering coincidence events to reconstruct the image and obtain the reconstructed image.

37. A digital device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the scattering correction method as described in any one of claims 1 to 32.

38. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the scattering correction method as described in any one of claims 1-32.

Images