A method and system for preprocessing reconstruction data of a cascaded gamma-photon coincidence imaging system

Abstract

This invention provides a data preprocessing method and system for reconstructed data in a cascaded gamma-photon coincidence imaging system, belonging to the field of nuclear medicine imaging technology. First, based on preset time windows, energy windows, and collimator types, the acquired events are initially screened to obtain an initial coincidence event set. Then, the coincidence events are back-projected, and parallel invalid events are eliminated by calculating the spatial geometric relationship between the back-projection line model and the back-projection plane. Second, the coordinates of the intersection points of the back-projection lines and the back-projection plane are calculated, and events whose intersection points are within the field of view are selected. Third, the probability of the voxel at the intersection point generating the gamma-photon pair is calculated, and events with a probability of zero are excluded. Fourth, the voxel count value corresponding to the intersection point is incremented by one, and the voxel number is used as the location marker for the event. Finally, a final screening is performed based on the voxel count value and a preset threshold. This invention, through a multi-level screening mechanism, eliminates random coincidence events, scattering coincidence events, and noisy data, significantly improving the image reconstruction quality of the system.

Description

Technical Field

[0001] This invention relates to the field of nuclear medicine imaging technology, and in particular to a method and system for preprocessing reconstruction data of a cascaded gamma-photon coincidence imaging system. Background Technology

[0002] In the field of nuclear medicine, theranostics is a rapidly developing cutting-edge research area in recent years. Its core principle is to achieve seamless integration of diagnosis and treatment through the same molecular target, typically employing the synergistic effect of radiolabeled imaging agents and therapeutic agents. This contrasts with traditional separate diagnostic and therapeutic approaches (such as using separate methods). 68 Ga for imaging and 177 Compared to Lu (therapeutic radionuclides), some radionuclides can emit both alpha / beta rays and gamma photons during their decay process. This allows for treatment using the alpha / beta rays produced during decay, while simultaneously enabling imaging using gamma photons, thus demonstrating significant potential for integrated diagnosis and treatment. Currently, gamma photon imaging primarily relies on positron emission tomography (PET) and single-photon emission computed tomography (SPECT). However, PET can only image back-to-back gamma photon pairs generated by conforming annihilation radiation, while traditional SPECT imaging systems suffer from inherent limitations such as low sensitivity, poor spatial resolution, and the inability to accurately locate decay sites through back projection. In clinical diagnosis and treatment, commonly used radionuclides (such as...) 177 Lu, 131 I and 111 The decay of In (etc.) simultaneously produces electrons and gamma photons. During the production of gamma photons, a cascade decay occurs, resulting in highly temporally and spatially correlated pairs of gamma photons (…). 177 Lu has approximately a 3.1% probability of generating cascaded gamma photon pairs ranging from 113 keV to 208 keV. Capitalizing on this characteristic, researchers developed cascaded gamma photon coincidence imaging technology, which achieves high spatial resolution and high signal-to-noise ratio imaging by directly locating the decay sites of radionuclides.

[0003] In cascaded gamma-photon coincidence imaging systems, cascaded gamma-photon pairs are typically identified by setting time and energy windows. However, this method can introduce random coincidence events, leading to incorrect decay location. Furthermore, gamma photons produced by radioactive decay undergo Compton scattering as they pass through the aperture collimator, slit collimator, and crystal detector, altering their trajectory. If scattered photons are mistaken for the original photons, further scattering coincidence events will be introduced, affecting the accuracy of decay location. On the other hand, to improve sensitivity, the geometric center of the aperture in the aperture collimator and the center point of the plane closest to the field of view in the slit collimator are often set to be the same distance from the system axis. However, this design may result in the straight line projected from the aperture collimator being parallel to the plane projected from the slit collimator, making it impossible to calculate their intersection. Summary of the Invention

[0004] To address the problems in the prior art, this invention provides a method and system for preprocessing reconstruction data in a cascaded gamma-photon coincidence imaging system. This invention effectively reduces random coincidence events, scattering coincidence events, and system noise through a multi-level event filtering mechanism, thereby significantly improving image reconstruction quality. To achieve the above objective, the technical solution is as follows:

[0005] On one hand, the present invention provides a method for preprocessing reconstruction data of a cascaded gamma-photon coincidence imaging system, the method comprising:

[0006] S1. Based on the system parameters of the cascaded γ-photon coincidence imaging system, the initial coincidence event set is obtained by comparing the preprocessed parameters with the preset time window threshold, energy window threshold and collimator type.

[0007] S2. Based on the initial set of coincidence events, identify and eliminate the events corresponding to when the back projection line is parallel to the back projection surface through the back projection line model and the back projection surface model, and obtain the second set of coincidence events.

[0008] S3. Based on the back-projection line and the back-projection surface corresponding to the second-stage coincidence event set, the third-stage coincidence event set is obtained by determining whether the coordinates of the intersection point of the back-projection line and the back-projection surface are within the preset field of view of the cascaded γ-photon coincidence imaging system.

[0009] S4. Based on the set of events that meet the third stage, by calculating the probability of generating γ photon pairs from the corresponding voxel decay in the field of view, events with a probability greater than zero are selected to obtain the set of events that meet the fourth stage.

[0010] S5. Based on the coordinates of the intersection point, the voxel count value and voxel number of the field of view are obtained through the calculation formula of voxel count and corresponding voxel number of the field of view.

[0011] S6. Based on the fourth stage matching event set, by traversing the voxel count value corresponding to the voxel number of the field of view, matching events below the preset voxel count threshold are removed to obtain the final matching event set.

[0012] Optionally, in S1, based on the system parameters of the cascaded gamma-photon coincidence imaging system, after preprocessing, an initial set of coincidence events is obtained by comparing them with preset time window thresholds, energy window thresholds, and collimator types, including:

[0013] S11. Based on the system parameters of the cascaded γ-photon coincidence imaging system, energy information, time information, and collimator type corresponding to the event are obtained through preprocessing.

[0014] S12. Based on the energy information, time information, and collimator type corresponding to the event, an initial set of matching events is obtained by comparing with preset time window thresholds, energy window thresholds, and collimator types.

[0015] Optionally, in S2, based on the initial set of coincidence events, events corresponding to when the back-projection line is parallel to the back-projection surface are identified and eliminated using the back-projection line model and the back-projection surface model, to obtain the second-stage set of coincidence events, including:

[0016] S21. Based on the initial set of coincident events, the back-projection line vector and the back-projection surface normal vector are obtained through the back-projection line model and the back-projection surface model.

[0017] S22. Based on the back-projection line vector and the back-projection surface normal vector, using the line-plane parallelism determination condition, identify and eliminate the events corresponding to when the back-projection line is parallel to the back-projection surface, and obtain the second stage conformity event set.

[0018] Optionally, S3, based on the back-projection line and the back-projection surface corresponding to the second-stage coincidence event set, determines whether the coordinates of the intersection point of the back-projection line and the back-projection surface are within the preset field of view of the cascaded γ-photon coincidence imaging system, and then filters to obtain the third-stage coincidence event set, including:

[0019] S31. Based on the back-projection line and the back-projection surface corresponding to the second stage conformity event set, the coordinates of the intersection point of the back-projection line and the back-projection surface are obtained through the line-plane intersection calculation formula.

[0020] S32. Based on the second-stage coincidence event set and the intersection point coordinates, the third-stage coincidence event set is obtained by determining whether the intersection point coordinates are within the preset field of view of the cascaded γ-photon coincidence imaging system.

[0021] Optionally, S5, based on the intersection point coordinates, uses a formula to calculate the voxel count and voxel number of the field of view, including:

[0022] S51. Based on the coordinates of the intersection point, the voxel count value of the field of view at the intersection point is increased by one to obtain the statistical voxel count value of the field of view;

[0023] S52. Based on the coordinates of the intersection point, obtain the voxel number of the field of view using the voxel numbering formula.

[0024] On the other hand, the present invention provides a preprocessing system for reconstruction data of a cascaded gamma-photon coincidence imaging system. This system is applied to a preprocessing method for reconstruction data of a cascaded gamma-photon coincidence imaging system. The system includes:

[0025] The event data acquisition module is used to filter the initial set of coincidence events by comparing them with preset time window thresholds, energy window thresholds and collimator types after preprocessing based on the system parameters of the cascaded gamma photon coincidence imaging system.

[0026] The second-stage filtering module is used to identify and remove events corresponding to when the back projection line is parallel to the back projection plane based on the initial set of matching events, using the back projection line model and the back projection plane model, to obtain the second-stage set of matching events.

[0027] The third-stage screening module is used to select the third-stage coincidence event set based on the back projection line and the back projection surface corresponding to the second-stage coincidence event set by determining whether the coordinates of the intersection point of the back projection line and the back projection surface are within the preset field of view of the cascaded γ-photon coincidence imaging system.

[0028] The fourth-stage screening module is used to select events with a probability greater than zero based on the third-stage matching event set by calculating the probability of generating γ-photon pairs from the corresponding field voxel decay, thus obtaining the fourth-stage matching event set.

[0029] The field-of-view voxel counting module is used to obtain the field-of-view voxel count value and the field-of-view voxel number based on the coordinates of the intersection point and the calculation formula of the field-of-view voxel count and the corresponding field-of-view voxel number.

[0030] The final filtering module is used to obtain the final set of matching events by traversing the voxel count value corresponding to the voxel number in the fourth stage and removing matching events below a preset voxel count threshold.

[0031] Compared with the prior art, the technical solution of the present invention has at least the following beneficial effects:

[0032] The above-mentioned solution employs three main methods. First, it establishes a dual geometric model of the back-projection line and surface and calculates their spatial relationship. Using a parallelism judgment mechanism, when the back-projection line is detected to be parallel to the back-projection surface, the system automatically identifies and eliminates invalid aperture coincidence events. Second, it calculates the spatial intersection point of the back-projection line and the back-projection surface. If the intersection point is determined to be outside the system's field of view, the aperture coincidence event is automatically filtered out. This technical solution effectively eliminates invalid aperture coincidence events in the reconstructed data caused by random coincidence events, scattering coincidence events, and other noise factors, where the back-projection intersection point is outside the field of view. Third, it calculates the probability of a nuclide decaying and generating a γ-photon pair in the voxel corresponding to the intersection point. When this probability is zero, the system automatically excludes the aperture coincidence event. This technical feature can further filter out zero-probability events caused by random coincidence, scattering coincidence and other noise interference; fourthly, by statistically analyzing the field-of-view voxel value corresponding to the back-projection intersection of the event, when the value is detected to be lower than a preset threshold, the slit coincidence event is removed, thus effectively filtering out random coincidence events, scattering coincidence events and other noise in the reconstructed data of the cascaded gamma-photon coincidence imaging system and improving the quality of the reconstructed data. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a flowchart of an embodiment of the preprocessing method for reconstructed data in a cascaded gamma-photon coincidence imaging system of the present invention;

[0035] Figure 2 A flowchart illustrating the initial set of coincidence events obtained by screening in an embodiment of the preprocessing method for reconstructed data from a cascaded gamma-photon coincidence imaging system of the present invention;

[0036] Figure 3 This is a schematic diagram of random coincidence in a cascaded gamma-photon coincidence imaging system;

[0037] Figure 4 This is a schematic diagram of Compton scattering occurring in the aperture and slit collimator of a cascaded gamma-photon coincidence imaging system.

[0038] Figure 5 This is a schematic diagram of Compton scattering occurring in the detector crystal of a cascaded gamma-photon coincidence imaging system.

[0039] Figure 6 A flowchart illustrating the screening process for obtaining the second-stage coincidence event set in an embodiment of the preprocessing method for reconstructed data from a cascaded γ-photon coincidence imaging system of the present invention.

[0040] Figure 7 This is a side view of the slit collimator in a cascaded gamma-photon coincidence imaging system;

[0041] Figure 8 This is a side view of the aperture collimator in a cascaded gamma-photon coincidence imaging system;

[0042] Figure 9 A flowchart illustrating the screening process for obtaining the third-stage coincidence event set in an embodiment of the preprocessing method for reconstructed data from a cascaded γ-photon coincidence imaging system of the present invention.

[0043] Figure 10 A flowchart illustrating the method for preprocessing reconstruction data of a cascaded γ-photon coincidence imaging system according to the present invention for obtaining voxel counts and voxel numbers in the field of view;

[0044] Figure 11 It is an image reconstructed from unprocessed reconstructed data using the MLEM algorithm iteratively.

[0045] Figure 12 It is an image reconstructed iteratively from preprocessed reconstructed data using the MLEM algorithm;

[0046] Figure 13 It is an image reconstructed from unprocessed reconstructed data using a direct backprojection algorithm;

[0047] Figure 14 It is an image reconstructed from preprocessed reconstructed data using a direct backprojection algorithm;

[0048] Figure 15 This is a system block diagram of an embodiment of the cascaded gamma-photon coincidence imaging system reconstruction data preprocessing system of the present invention.

[0049] Figure labels: 1. Trajectory of a γ-photon after Compton scattering on the aperture collimator; 2. Trajectory of a γ-photon through the aperture collimator without Compton scattering; 3. Trajectory of a γ-photon through the slit collimator without Compton scattering; 4. Aperture collimator; 5. Slit collimator; 6. Location of nuclide decay; 7. Crystal detector; 8. Field of view of the cascaded γ-photon coincidence imaging system; 9. Trajectory of a γ-photon after Compton scattering on the crystal; 11. Center point of the slit collimator on the side plane closest to the field of view; 12. Geometric center of the aperture in the aperture collimator; 13. Circular surface formed at the narrowest point of the aperture collimator; 14. Axis of the aperture in the aperture collimator; 15. First trajectory of the cascaded γ-photon pair generated after nuclide decay; 101. Second trajectory of the cascaded γ-photon pair generated after nuclide decay; 102. Detailed Implementation

[0050] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0051] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0052] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0053] like Figure 1 The flowchart shown is an embodiment of the preprocessing method for reconstructed data of a cascaded gamma-photon coincidence imaging system according to the present invention. The present invention provides a preprocessing method for reconstructed data of a cascaded gamma-photon coincidence imaging system, which is implemented by a preprocessing system for reconstructed data of a cascaded gamma-photon coincidence imaging system. The method includes:

[0054] S1. Based on the system parameters of the cascaded γ-photon coincidence imaging system, the initial coincidence event set is obtained by comparing the preprocessed parameters with the preset time window threshold, energy window threshold and collimator type.

[0055] Specifically, such as Figure 2 The flowchart for obtaining the initial set of matching events through filtering includes:

[0056] S11. Based on the system parameters of the cascaded γ-photon coincidence imaging system, energy information, time information, and collimator type corresponding to the event are obtained through preprocessing.

[0057] S12. Based on the energy information, time information, and collimator type corresponding to the event, an initial set of matching events is obtained by comparing with preset time window thresholds, energy window thresholds, and collimator types.

[0058] Furthermore, in an optional embodiment, event information of the interaction between γ photons and the detection unit within the cascaded γ photon coincidence imaging system is obtained, specifically including the energy information, time information, and collimator type corresponding to the event. The collimator type includes aperture collimator 5 and slit collimator 6.

[0059] Event information can be obtained through a cascaded gamma-photon coincidence imaging system or through Monte Carlo simulation using software such as GATE or GEANT4. Specifically, for physical event acquisition, energy information is calculated by integrating the amplitude of the signal pulse over time; time information is determined by the trigger time of the signal pulse captured by the data acquisition module within the cascaded gamma-photon coincidence imaging system; and the detector ring number, module number on the ring, and collimator type can be directly obtained from the module interacting with the gamma photon. For event information acquired using the Monte Carlo simulation method with GATE software, the above information can be directly obtained. Since the cascaded gamma-photon coincidence imaging system is not yet fully constructed, this embodiment uses the Monte Carlo simulation method with GATE software to acquire event data. However, it is understood that this embodiment does not limit the selection of the data acquisition method. Furthermore, in this embodiment, heat source columns with radii of 0.5, 0.6, 0.75, 0.9, 1, and 1.25 mm were used for simulation with GATE settings, the activity of the radioactive source was 1 mCi, and the simulation time was set to 30,000 s. This embodiment does not limit the settings of the radius of the heat source column, the activity of the radioactive source, or the simulation time.

[0060] When using GATE software to collect event data using the Monte Carlo simulation method, random coincidence events and Compton scattering events in real systems can be simulated. In an optional embodiment, a schematic diagram of a random coincidence event is shown below. Figure 3 As shown, dashed arrows represent random coincidence events, and solid arrows represent true coincidence events. In an optional embodiment, a schematic diagram of Compton scattering occurring in the collimator is shown below. Figure 4 As shown in the figure, the nuclide decay position 7 is illustrated. The trajectory 1 of the γ photon after Compton scattering at the aperture collimator, the trajectory 4 of the γ photon after Compton scattering at the slit collimator, the trajectory 2 of the γ photon passing through the aperture collimator without Compton scattering, and the trajectory 3 of the γ photon passing through the slit collimator without Compton scattering are also shown. In an optional embodiment, a schematic diagram of Compton scattering occurring in the detector crystal is shown below. Figure 5 As shown, the first trajectory 101 of the cascaded γ-photon pairs generated after the decay of the nuclide, the second trajectory 102 of the cascaded γ-photon pairs generated after the decay of the nuclide, and the trajectory 11 of the γ-photon generated after Compton scattering in the crystal detector 8 and Compton scattering on the crystal.

[0061] Furthermore, the initial screening of matching events is mainly carried out through energy windows and time windows, and each screening must involve one γ photon passing through the aperture collimator 5 and another passing through the slit collimator 6, and a pair of matching events is considered a valid match;

[0062] The time window meets the following filtering formula:

[0063]

[0064] In the formula, t 1 represents the timestamp of event 1. t 2 represents the timestamp of event 2. The time difference between the detection of events 1 and 2; if If the window size is smaller than the set window size, then this event timing is correct.

[0065] The above The value can be set according to requirements; in this embodiment... The value is set to 8 ns. Of course, it is understood that this embodiment is not limited to this. Setting the value;

[0066] The energy window meets the screening formula as follows:

[0067]

[0068] In the formula, E 1 represents the energy value of event 1. E 2 represents the energy value of event 2. E low,down The lower energy threshold corresponding to the low-energy gamma photons produced by the decay of nuclides. E low,up The upper threshold of the energy window corresponding to the low-energy gamma photons produced by the decay of nuclides. E high,down The lower threshold of the energy window corresponding to the high-energy gamma photons produced by the decay of nuclides. E high,up The threshold of the energy window corresponding to the high-energy γ photon produced by the decay of nuclide; if formula (2) holds, then this pair of events is energetically compatible.

[0069] Furthermore, the above-mentioned types of nuclides are 177 Lu, 131 I and 111 In, etc., this embodiment adopts... 177 As Lu is a radioactive source, it is understandable that this embodiment does not limit the selection of nuclides. Because... 177 The decay of Lu nuclides produces a pair of gamma photons with energies of 113 keV and 208 keV. The energy window of these low-energy gamma photons... E low,down and E low,up And the energy window of high-energy gamma photons E high,down and E high,up The values ​​are set according to 113 keV and 208 keV respectively, in this embodiment. Elow,down , E low,up , E high,down , E high,up The values ​​were set to 100 keV, 127 keV, 191 keV and 225 keV respectively. Of course, it is understood that this embodiment does not limit the setting of the energy window.

[0070] Based on the aforementioned time window, energy window, and collimator type, 397,301 pairs of aperture matching events were initially screened, of which random matching accounted for approximately 9.27%.

[0071] S2. Based on the initial set of coincidence events, identify and eliminate the events corresponding to when the back projection line is parallel to the back projection surface through the back projection line model and the back projection surface model, and obtain the second set of coincidence events.

[0072] Specifically, such as Figure 6 This is a flowchart of the process for obtaining the second-stage set of matching events in this invention, including:

[0073] S21. Based on the initial set of coincident events, the back-projection line vector and the back-projection surface normal vector are obtained through the back-projection line model and the back-projection surface model.

[0074] S22. Based on the back-projection line vector and the back-projection surface normal vector, the event corresponding to the parallelism between the back-projection line and the back-projection surface is identified and eliminated using the vector calculation formula for line-plane parallelism, thus obtaining the second-stage conformity event set.

[0075] Furthermore, the coordinates of the corresponding crystal unit center point are calculated based on the initially selected matching events and crystal action positions; then, back projection modeling is performed based on the center point of the hole collimator 5 and the center of the slit collimator 6.

[0076] In an optional embodiment, the back-projection line is modeled based on the coordinates of the center point of the aperture collimator 5 and the corresponding γ-photon action position:

[0077]

[0078]

[0079] In the formula, The coordinates of the geometric center of the detector unit corresponding to the event of passing through the collimator 5 and the interaction of the crystal detector 8 module are given. The three-dimensional coordinates of the center point of the hole in the hole collimator 5 corresponding to this event. The distance between the center point of the aperture collimator 5 and the geometric center point of the detector unit. The unit vector pointing from the geometric center of the detector unit to the geometric center of the aperture collimator 5.

[0080] Hole collimator 5 Figure 8 As shown, the geometric center point of the hole in the hole collimator 5 is the geometric center 13 of the hole in the hole collimator. The coordinates of the geometric center 13 of the hole in the hole collimator can be set according to requirements. Specifically, in this embodiment, the distance from the geometric center 13 of each hole in the hole collimator to the axis of the system is 30 mm.

[0081] Furthermore, in an optional embodiment, a back-projection is performed based on the coordinates of the center point of the slit collimator 6 and the corresponding γ-photon action position to model the back-projection surface:

[0082]

[0083]

[0084] In the formula, The coordinates of the geometric center of the detector unit corresponding to the event of passing through the slit collimator 6 and interacting with the crystal detector 8 are given. The three-dimensional coordinates are those of the center point of the plane of the field of view region 9 of the cascaded gamma-photon coincidence imaging system near the slit collimator 6 corresponding to the event. The distance between the center point of the slit collimator 6 and the geometric center point of the detector unit. The unit vector pointing from the geometric center of the detector unit to the geometric center of the slit collimator 6;

[0085] 6-piece sewing collimator Figure 7 As shown, the center point of the plane of the slit collimator 6 near the field of view 9 of the cascaded γ-photon coincidence imaging system is the center point 12 of the side plane of the slit collimator near the field of view. The geometric center position of the slit collimator 6 can be set as needed. In this embodiment, the center point 12 of the side plane of each slit collimator near the field of view is 30 mm away from the axis of the system.

[0086] In the entire cascaded gamma-photon coincidence imaging system, the plane at the narrowest point of the slit collimator 6 is parallel to z Since the axis is defined, another vector in the back projection plane can be found as:

[0087] (7)

[0088] Therefore, based on the two unrelated vectors in the back projection plane and (They must not be collinear), calculate the unit normal vector of the plane using the cross product:

[0089]

[0090] In the formula, It is the unit normal vector of the back-projected surface.

[0091] When a line is parallel to a plane, the dot product of the line direction vector and the plane normal vector is zero. By checking whether the dot product is zero, events corresponding to line-plane parallelism can be eliminated.

[0092] like If the back-projection plane is parallel to the back-projection line, the matching event should be discarded, and finally the second-stage matching event set is obtained.

[0093] S3. Based on the back-projection line and the back-projection surface corresponding to the second-stage coincidence event set, the third-stage coincidence event set is obtained by determining whether the coordinates of the intersection point of the back-projection line and the back-projection surface are within the preset field of view of the cascaded γ-photon coincidence imaging system.

[0094] Specifically, such as Figure 9 This is a flowchart of the process for obtaining the third-stage matching event set in this invention, including:

[0095] S31. Based on the back-projection line and the back-projection surface corresponding to the second stage conformity event set, the coordinates of the intersection point of the back-projection line and the back-projection surface are obtained through the line-plane intersection calculation formula.

[0096] S32. Based on the second-stage coincidence event set and the intersection point coordinates, the third-stage coincidence event set is obtained by determining whether the intersection point coordinates are within the preset field of view of the cascaded γ-photon coincidence imaging system.

[0097] Furthermore, the parameters of the back-projection plane equation are calculated based on the normal vector of the back-projection plane and the center point of the slit collimator 6, as shown in equations (9) and (10) below:

[0098]

[0099]

[0100] In the formula, Let be the parameters of the equation of the back projection plane, where with vector Equal, and D It is calculated using formula (10);

[0101] When the back-projection plane is not parallel to the back-projection line, the intersection of the back-projection line and the back-projection plane is calculated as shown in equations (11) and (12) below:

[0102]

[0103]

[0104] In the formula, t This is the distance from point 13, the geometric center of the hole in the hole collimator, to the intersection point (intersection point: the intersection of the back projection line and the back projection plane). for , The coordinates of the intersection point are 3D coordinates.

[0105] Furthermore, by determining the coordinates of the intersection point Whether it is within the system's preset field of view. If the intersection point coordinates are outside the field of view, the matching data is removed, and the third stage matching event set is obtained.

[0106] S4. Based on the set of events that meet the third stage, by calculating the probability of generating γ photon pairs from the corresponding voxel decay in the field of view, events with a probability greater than zero are selected to obtain the set of events that meet the fourth stage.

[0107] Specifically, the probability of the cascaded γ-photon pair generated by the decay of the voxel nucleus corresponding to the intersection point being the coincidence event of the aperture is calculated using the formula shown in (14):

[0108] (14)

[0109] In the formula, The probability of the coincidence event is the cascaded γ-photon pairs generated by the decay of the voxel nucleus. This represents the probability that a gamma photon generated by the decay of the voxel nucleus will be detected by the crystal detector 8 after passing through the aperture collimator 5. The probability that a γ photon generated by the decay of the voxel nucleus is detected by the crystal detector 8 after passing through the slit collimator 6;

[0110] Furthermore, the probability that the γ photon generated by the voxel decay will be detected by the crystal detector 8 after passing through the aperture collimator 5 is shown in the following equation (15):

[0111]

[0112] In the formula, The attenuation coefficient of γ photons for aperture collimator 5. The length of the intersection between the trajectory of the γ-photon and the aperture collimator 5. denoted as the attenuation coefficient of the γ photon as it passes through crystal detector 8. The length of the γ photon as it passes through crystal detector 8. The solid angle formed by the trajectory of the γ-photon and the crystal unit;

[0113] Furthermore, the collimator is made of any one of high-density metals such as tungsten or lead. Specifically, this embodiment uses tungsten metal as the collimator; however, it is understood that this embodiment is not limited to the selection of the collimator material. In addition, since this embodiment of the invention uses... 177 Since Lu is the radiation source, the attenuation coefficient for a 208 keV gamma photon passing through the collimator is... The attenuation coefficient for a 113 keV γ photon passing through a collimator is: ;

[0114] Since the collimator's function is to block gamma photons and it has a high density, therefore... If the threshold is exceeded, it can be The overall threshold is set to zero. Optionally, this threshold can be set as needed. Specifically, in this embodiment, the penetration distance at which the collimator's attenuation rate for γ-photons reaches 95% is set as the threshold. It is understood that this embodiment does not limit the selection of the collimator's blocking capability for γ-photons. Therefore, when a γ-photon with an energy of 208 keV passes through this collimator, if... Then The overall value is set to zero; when a γ photon with energy of 113 keV passes through this collimator, if... Then Set the total to zero;

[0115] The crystal detector 8 is made of any one of the following: lanthanum bromide (LaBr3), bismuth germanate (BGO), lutetium silicate (LSO), yttrium lutetium silicate (LYSO), and sodium iodide (NaI:Tl). Specifically, this embodiment uses lanthanum bromide crystal as the detector material; however, it is understood that this embodiment does not limit the selection of the detector material. Therefore, the attenuation coefficient for a 208 keV γ photon passing through the crystal detector 8 is... =0.128; the attenuation coefficient for a 113 keV γ photon passing through crystal detector 8 is =0.478;

[0116] Solid angle The calculation formula is:

[0117] (16)

[0118] In the formula, S is the lateral area of ​​the field of view region 9 of the crystal detector 8 near the cascaded γ-photon coincidence imaging system. The angle between the incident direction of the γ-photon through the collimator 5 and the crystal axis. Let be the distance from the center point of voxel 9 in the field of view of the cascaded gamma-photon coincidence imaging system to the center of crystal detector 8, where:

[0119]

[0120] Since the continuous crystal module is discretized into a 50×50 detector unit array with a size of 1 mm × 1 mm × 6 mm in S2 of this embodiment, therefore ;

[0121] Furthermore, the probability that the γ photon generated by the voxel decay will be detected by the crystal detector 8 after passing through the slit collimator 6 is shown in the following equation (17):

[0122]

[0123] In the formula, The attenuation coefficient of γ photons for the slit collimator 6 is given. The length of the intersection between the trajectory of the γ-photon and the slit collimator 6. denoted as the attenuation coefficient of the γ photon as it passes through crystal detector 8. The length of the γ photon as it passes through crystal detector 8. The solid angle formed by the trajectory of the γ-photon and the crystal unit;

[0124] In the above formula, , , The parameter settings are the same as those for the orifice collimator 5.

[0125] Since the inner side of the aperture collimator 5 is composed of two conical surfaces, the distance the γ photon travels through the aperture collimator 5 is calculated. Iterative solutions are required, which consumes a significant amount of time. Therefore, this embodiment addresses this issue. The calculation is simplified. First, determine whether the angle between the trajectory of the gamma photon and the axis of the aperture collimator is less than the semi-apex angle of the aperture's cone. The axis of aperture collimator 5 is as follows: Figure 8 The axis of the central aperture collimator is shown as 15. Next, the intersection of the gamma photon's trajectory and the plane at the narrowest point of the aperture collimator is calculated, and it is determined whether the intersection point is smaller than the radius of the inner aperture diameter at the narrowest point of the aperture collimator. The narrowest point of the aperture collimator is a circle, as shown in the schematic diagram. Figure 8 The circular surface 14 formed at the narrowest point of the central aperture collimator is shown. Ultimately, if the angle between the trajectory of the γ-photon and the axis of the aperture collimator is less than the semi-apex angle of the aperture, and the distance from the intersection of the trajectory and the plane at the narrowest point of the aperture collimator to the geometric center of aperture collimator 5 is less than the radius of the inner aperture diameter at the narrowest point of the aperture collimator, then it is considered that the γ-photon can pass through aperture collimator 5. ;on the contrary, It equals positive infinity, that is ;

[0126] The semi-apex angle of the cone corresponding to the hole in the collimator can be set as needed. Specifically, in this embodiment, the semi-apex angle of the cone corresponding to the hole is an acute angle with a tangent value of 0.25. Of course, it is understood that this embodiment does not limit the selection of the semi-apex angle of the cone corresponding to the hole.

[0127] The radius of the inner diameter at the narrowest point of the hole alignment can be set according to requirements. Specifically, in this embodiment, the radius of the inner diameter at the narrowest point of the hole alignment is 0.5 mm. Of course, it is understood that this embodiment does not limit the selection of the inner diameter radius at the narrowest point of the hole alignment.

[0128] To increase the sensitivity of the field of view, the axes of all aperture collimators 5 need to be oriented towards the field of view region 9 of the cascaded gamma-photon coincidence imaging system, thus resulting in an angle between the axes and the positive z-axis. The angle between the axis of the aperture in the aperture collimator 5 and the positive z-axis can be set as needed. Specifically, in this embodiment, the angles between the axis of the aperture in each aperture collimator 5 and the positive z-axis are set to 55.448°, 77.42°, 102.58°, and 124.552° respectively. It is understood that this embodiment does not limit the selection of the angle between the axis of the aperture in the aperture collimator 5 and the positive z-axis.

[0129] Solid angle The calculation formula is:

[0130] (18)

[0131] In the formula, The angle between the incident direction of the γ-photon through the slit collimator 6 and the crystal axis. Let be the distance from the center point of the 9 voxel in the field of view of the cascaded gamma-photon coincidence imaging system to the center of the 8-cell crystal detector, where:

[0132]

[0133] Determine the calculated probability Is it zero? If it is zero, then remove the matching event.

[0134] By judging the calculated probability Whether it is zero, the total number of hole gap matching events that were eliminated is 66318.

[0135] S5. Based on the coordinates of the intersection point, the voxel count value and voxel number of the field of view are obtained through the calculation formula of voxel count and corresponding voxel number of the field of view.

[0136] Specifically, such as Figure 10 The flowchart for obtaining the voxel count and voxel number of the field of view according to the present invention includes:

[0137] S51. Based on the coordinates of the intersection point, the voxel count value of the field of view at the intersection point is increased by one to obtain the statistical voxel count value of the field of view;

[0138] S52. Based on the coordinates of the intersection point, obtain the voxel number of the field of view using the voxel numbering formula.

[0139] Furthermore, the intersection coordinates are converted into field-of-view voxel numbers using the following formula:

[0140]

[0141] In the formula, , , These represent the lengths of the field of view region 9 of the cascaded gamma-photon coincidence imaging system along the x, y, and z axes, respectively. , , The voxels of the field of view are along x , y , z Shaft dimensions, The voxels corresponding to the intersection points are respectively in x , y , z The corresponding voxel number on the axis.

[0142] S6. Based on the fourth stage matching event set, by traversing the voxel count value corresponding to the voxel number of the field of view, matching events below the preset voxel count threshold are removed to obtain the final matching event set.

[0143] Specifically, based on the location number of the event Find the field voxel corresponding to the back-projection intersection; then, determine whether the field voxel value is less than the set threshold. C_th If the value is less than the threshold, the matching event is deleted; otherwise, it is retained.

[0144] Furthermore, the aforementioned threshold C_th The threshold value can be set according to the voxel size of the field of view; in this embodiment, it is used as an example. C_th Set it to 5;

[0145] Based on whether the voxel count value corresponding to the judgment event exceeds the threshold, the number of hole gap matching events that can be removed again is 55565.

[0146] Furthermore, based on the above steps S1-S6, the initial screening matching events in S1 can be reduced to 211,132 pairs, of which random matching events are reduced by 94.29%. The matching events without preprocessing and the matching events after preprocessing are then subjected to iterative reconstruction using the MLEM algorithm and direct back-projection reconstruction, respectively. Figure 11It is an image reconstructed from unprocessed reconstructed data using the MLEM algorithm iteratively. Figure 12 It is an image reconstructed iteratively from preprocessed reconstructed data using the MLEM algorithm; Figure 13 It is an image reconstructed from unprocessed reconstructed data using a direct backprojection algorithm; Figure 14 The images are reconstructed from preprocessed reconstructed data using a direct backprojection algorithm. As can be seen from the images above, the image noise after preprocessing according to this invention, whether reconstructed iteratively using the MLEM algorithm or through direct backprojection, is lower than that of images directly reconstructed from unprocessed coincidence data. Therefore, the reconstructed data preprocessing method of this invention can effectively reduce random coincidence events, scattering coincidence events, and other noise in the reconstructed data of a cascaded gamma-photon coincidence imaging system, significantly improving image reconstruction quality.

[0147] like Figure 15 The diagram shown is a system block diagram of an embodiment of the cascaded gamma-photon coincidence imaging system reconstruction data preprocessing system of the present invention. The present invention provides a cascaded gamma-photon coincidence imaging system reconstruction data preprocessing system, which is applied to a cascaded gamma-photon coincidence imaging system reconstruction data preprocessing method. The system includes:

[0148] The event data acquisition module is used to filter the initial set of coincidence events by comparing them with preset time window thresholds, energy window thresholds and collimator types after preprocessing based on the system parameters of the cascaded gamma photon coincidence imaging system.

[0149] The second-stage filtering module is used to identify and remove events corresponding to when the back projection line is parallel to the back projection plane based on the initial set of matching events, using the back projection line model and the back projection plane model, to obtain the second-stage set of matching events.

[0150] The third-stage screening module is used to select the third-stage coincidence event set based on the back projection line and the back projection surface corresponding to the second-stage coincidence event set by determining whether the coordinates of the intersection point of the back projection line and the back projection surface are within the preset field of view of the cascaded γ-photon coincidence imaging system.

[0151] The fourth-stage screening module is used to select events with a probability greater than zero based on the third-stage matching event set by calculating the probability of generating γ-photon pairs from the corresponding field voxel decay, thus obtaining the fourth-stage matching event set.

[0152] The field-of-view voxel counting module is used to obtain the field-of-view voxel count value and the field-of-view voxel number based on the coordinates of the intersection point and the calculation formula of the field-of-view voxel count and the corresponding field-of-view voxel number.

[0153] The final filtering module is used to obtain the final set of matching events by traversing the voxel count value corresponding to the voxel number in the fourth stage and removing matching events below a preset voxel count threshold.

[0154] This invention provides a data preprocessing method and system for reconstructed data in a cascaded gamma-photon coincidence imaging system, belonging to the field of nuclear medicine imaging technology. First, based on preset time windows, energy windows, and collimator types, the acquired events are initially screened to obtain an initial coincidence event set. Then, the coincidence events are back-projected, and parallel invalid events are eliminated by calculating the spatial geometric relationship between the back-projection line model and the back-projection plane. Next, the coordinates of the intersection points of the back-projection lines and the back-projection plane are calculated, and events whose intersection points are within the field of view are selected. Third, the probability of the voxel at the intersection point generating the gamma-photon pair is calculated, and events with a probability of zero are excluded. Fourth, the voxel count value corresponding to the intersection point is incremented by one, and the voxel number is used as the location marker for the event. Finally, a final screening is performed based on the voxel count value and a preset threshold. This invention, through a multi-level screening mechanism, eliminates random coincidence events, scattering coincidence events, and noisy data, significantly improving the image reconstruction quality of the system.

[0155] It is understood that the present invention has been described through the above embodiments and should not be construed as limiting the implementation and scope of the present invention. Those skilled in the art will recognize that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the present invention. Furthermore, under the teachings of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of the present invention.

Claims

1. A method for preprocessing reconstruction data of a cascaded gamma-photon coincidence imaging system, characterized in that, The method includes: S1. Based on the system parameters of the cascaded γ-photon coincidence imaging system, the initial coincidence event set is obtained by comparing the preprocessed parameters with the preset time window threshold, energy window threshold and collimator type. S2. Based on the initial set of coincident events, identify and eliminate events corresponding to when the back projection line is parallel to the back projection surface using the back projection line model and the back projection surface model, to obtain the second set of coincident events. S3. Based on the back-projection line and the back-projection surface corresponding to the second stage coincidence event set, the third stage coincidence event set is obtained by determining whether the coordinates of the intersection point of the back-projection line and the back-projection surface are within the preset field of view of the cascaded γ-photon coincidence imaging system. S4. Based on the set of events that meet the third stage criteria, by calculating the probability of generating γ photon pairs from the corresponding voxel decay in the field of view, events with a probability greater than zero are selected to obtain the set of events that meet the fourth stage criteria. S5. Based on the intersection coordinates, the field of view voxel count value and field of view voxel number are obtained through the calculation formula of field of view voxel count and corresponding field of view voxel number. S6. Based on the fourth stage matching event set, by traversing the field of view voxel count value corresponding to the field of view voxel number, matching events below the preset field of view voxel count threshold are removed to obtain the final matching event set.

2. The data preprocessing method for reconstructed data of the cascaded gamma-photon coincidence imaging system according to claim 1, characterized in that, In step S1, based on the system parameters of the cascaded gamma-photon coincidence imaging system, after preprocessing, the initial set of coincidence events is obtained by comparing them with preset time window thresholds, energy window thresholds, and collimator types. This set includes: S11. Based on the system parameters of the cascaded γ-photon coincidence imaging system, energy information, time information, and collimator type corresponding to the event are obtained through preprocessing. S12. Based on the energy information, time information, and collimator type corresponding to the event, an initial set of matching events is obtained by comparing with preset time window threshold, energy window threshold, and collimator type.

3. The data preprocessing method for reconstructed data of the cascaded gamma-photon coincidence imaging system according to claim 1, characterized in that, In step S2, based on the initial set of coincidence events, events corresponding to when the back-projection line is parallel to the back-projection surface are identified and eliminated using the back-projection line model and the back-projection surface model, resulting in a second set of coincidence events, including: S21. Based on the initial set of coincident events, obtain the back-projection line vector and the back-projection surface normal vector through the back-projection line model and the back-projection surface model. S22. Based on the back-projection line vector and the back-projection surface normal vector, and using the line-plane parallelism determination condition, identify and eliminate the events corresponding to when the back-projection line is parallel to the back-projection surface, to obtain the second stage conformity event set.

4. The data preprocessing method for reconstructed data of the cascaded gamma-photon coincidence imaging system according to claim 1, characterized in that, S3, based on the back-projection line and the back-projection surface corresponding to the second-stage coincidence event set, determines whether the coordinates of the intersection point of the back-projection line and the back-projection surface are within the preset field of view of the cascaded γ-photon coincidence imaging system, and then filters to obtain the third-stage coincidence event set, including: S31. Based on the back-projection line and the back-projection surface corresponding to the second stage conforming event set, the coordinates of the intersection point of the back-projection line and the back-projection surface are obtained by using the line-plane intersection calculation formula; S32. Based on the second stage coincidence event set and the intersection point coordinates, the third stage coincidence event set is obtained by determining whether the intersection point coordinates are within the preset field of view of the cascaded γ-photon coincidence imaging system.

5. The data preprocessing method for reconstructed data of a cascaded gamma-photon coincidence imaging system according to claim 1, characterized in that, S5, based on the intersection point coordinates, uses a formula to calculate the voxel count and voxel number of the field of view, including: S51. Based on the intersection point coordinates, the field of view voxel value at the intersection point coordinates is incremented by one to obtain the statistical field of view voxel count value; S52. Based on the intersection coordinates, obtain the field of view voxel number using the field of view voxel number calculation formula.

6. A preprocessing system for reconstructed data of a cascaded gamma-photon coincidence imaging system, used to implement the preprocessing method for reconstructed data of a cascaded gamma-photon coincidence imaging system as described in any one of claims 1-5, characterized in that, The system includes: The event data acquisition module is used to filter the initial set of coincidence events by comparing them with preset time window thresholds, energy window thresholds and collimator types after preprocessing based on the system parameters of the cascaded gamma photon coincidence imaging system. The second-stage filtering module is used to identify and remove events corresponding to when the back projection line is parallel to the back projection surface based on the initial set of matching events, using the back projection line model and the back projection surface model, to obtain the second-stage set of matching events. The third-stage screening module is used to filter out the third-stage coincidence event set based on the back-projection line and the back-projection surface corresponding to the second-stage coincidence event set by determining whether the coordinates of the intersection point of the back-projection line and the back-projection surface are within the preset field of view of the cascaded γ-photon coincidence imaging system. The fourth-stage screening module is used to filter out events with a probability greater than zero by calculating the probability of generating γ-photon pairs from the corresponding field-of-view voxel decay based on the third-stage matching event set, thereby obtaining the fourth-stage matching event set. The field of view voxel counting module is used to obtain the field of view voxel count value and the field of view voxel number based on the intersection coordinates and the calculation formula of field of view voxel count and corresponding field of view voxel number. The final filtering module is used to obtain the final set of matching events by traversing the field of view voxel count value corresponding to the field of view voxel number and removing matching events below a preset field of view voxel count threshold based on the fourth stage matching event set.

Images