True event discrimination method, device, system and equipment and storage medium
By screening and judging the detection data in the PET system, using the conformity time window and Compton cone to identify the true event, the false conformity event problem caused by Compton scattering is solved, and the quality of image reconstruction is improved.
Patent Information
- Application Number
- CN202311546117.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-20
- Publication Date
- 2025-05-20
AI Technical Summary
In existing PET systems, gamma photons of annihilation events may experience Compton scattering through human tissue or scintillation crystals before incident on the detector, resulting in the acquired matching events not true matching events, affecting the image reconstruction effect.
By obtaining detection data, including location information and time information, setting the time window and energy window, filtering out the detection data within the time window, and determining whether the detection data corresponds to the true event based on the energy window and Compton cone.
提高了真事件甄别的准确度，有效利用了低能真事件的数据，提升了图像重建的质量和分辨率。
Smart Images
Figure CN120020882A_ABST
Abstract
Claims
1. A true event identification method, characterized in that: The true event identification method comprises: Acquire detection data, where the detection data includes location information and time information; A time window and an energy window are set to meet the requirements, and the detection data in the time window are screened out based on the relationship between the energy sums corresponding to the two groups of detection data and the upper energy threshold of the energy window, so that only two groups of detection data are included in the time window and the energy sums of the two groups of detection data do not exceed the upper energy threshold; For the two groups of detection data that are selected to be within the time window, based on the energy window and whether the two groups of detection data respectively correspond to the Compton cone, it is determined whether each group of detection data corresponds to a true event.
2. The true event identification method according to claim 1, characterized in that: Acquiring detection data includes: Acquire a scintillation pulse signal; The scintillation pulse signal is sampled to obtain the detection data, where the detection data includes time-amplitude pairs.
3. The true event identification method according to claim 2, characterized in that: The scintillation pulse signal is acquired based on a detection module, and the location information of the detection data includes an IP address code of the detection module and a CH address code of each scintillation crystal in a scintillation crystal array of the detection module.
4. The true event identification method according to claim 2, characterized in that: Sampling is performed using a multi-voltage threshold sampling method or an ADC sampling method.
5. The true event identification method according to claim 1, characterized in that: The compliance time window and energy window are determined according to prior information.
6. The true event identification method according to claim 1, characterized in that: Setting a time window and an energy window that meet the requirements, and filtering out detection data that meet the time window based on the energy corresponding to each of the two groups of detection data and the relationship with the upper energy threshold, including: The detection data are screened using the matching time windows to obtain the detection data within a number of matching time windows; Filter the detection data in the same time window based on the location information of the detection data to obtain the detection data that meets the location requirements; Filter out the detection data that meets the time window, and only include two groups of detection data in the time window; An energy window is set, and based on the relationship between the energy corresponding to each of the two groups of detection data and the upper energy threshold of the energy window, two groups of detection data with corresponding energy within the same time window and not exceeding the upper energy threshold are screened out.
7. The true event identification method according to claim 6, characterized in that: Filter the detection data in the same matching time window based on the location information of the detection data, including: A group of detection data coded by the IP address belonging to the same detection module is screened, and the group of detection data is divided into a number of detection data subsets based on the CH address code of the scintillation crystal.
8. The true event identification method according to claim 6, characterized in that: The matching time window only includes two groups of detection data, wherein the two groups of detection data are located in the same matching time window and only correspond to the IP address codes of two detection modules.
9. The true event identification method according to claim 6, characterized in that: Based on the relationship between the energy corresponding to the two sets of detection data and the upper energy threshold of the energy window, two sets of detection data with the corresponding energy within the same time window and not exceeding the upper energy threshold are screened out, including: Two groups of detection data within the same conforming time window are screened based on the relationship between the energy sum corresponding to each group of detection data and the upper limit energy threshold of the energy window: if the energy sum corresponding to at least one group of the two groups of detection data exceeds the upper limit energy threshold, the detection data within the conforming time window is discarded; if the energy sum corresponding to each group of detection data does not exceed the upper limit energy threshold, the detection data within the conforming time window is retained.
10. The true event identification method according to claim 9, characterized in that: The energy sum corresponding to each group of detection data is the sum of the deposition energies corresponding to each detection data subset included in the group of detection data, and the deposition energy corresponding to each detection data subset is obtained based on the integration of the pulse waveform, and the pulse waveform is obtained by fitting each detection data subset; the detection data subset is a data set corresponding to the CH address encoding of a scintillation crystal in a detection module.
11. The true event identification method according to claim 1, characterized in that: For the two groups of detection data that are selected and meet the time window, based on the energy window and whether the two groups of detection data correspond to Compton cones respectively, it is judged whether each group of detection data corresponds to a true event, including: Based on the determination result of the energy window, it is determined whether to continue to determine whether the two sets of detection data correspond to true events based on whether the two sets of detection data respectively correspond to Compton cones.
12. The true event identification method according to claim 11, characterized in that: Based on the determination result of the energy window, it is determined whether to continue to determine whether each set of detection data corresponds to a true event based on whether the two sets of detection data correspond to Compton cones, including: Based on the energy window alone, it can be determined that both sets of detection data correspond to true events, without further determining whether each set of detection data corresponds to a true event based on whether the two sets of detection data each correspond to a Compton cone. It is impossible to determine whether the two sets of detection data correspond to true events based on the energy window alone. We continue to determine whether the two sets of detection data correspond to true events based on whether the two sets of detection data each correspond to the Compton cone.
13. The true event identification method according to any one of claims 1, 11-12, characterized in that: Based on the energy window and whether the two sets of detection data correspond to Compton cones, it is determined whether each set of detection data corresponds to a low-energy true event, including: Set an energy window and determine whether the two sets of detection data correspond to true events based on the energy window: If the energy sums corresponding to the two sets of detection data are both within the energy window, then it is determined that the two sets of detection data correspond to true events; If at least one of the energy sums corresponding to the two sets of detection data is outside the energy window, it is impossible to determine whether the two sets of detection data correspond to true events. Continue to determine whether the two sets of detection data correspond to true events based on the Compton cones corresponding to the two sets of detection data.
14. The true event identification method according to claim 13, characterized in that: Based on the Compton cones corresponding to the two sets of detection data, determining whether the two sets of detection data both correspond to true events includes: Based on the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data, it is determined whether the two sets of detection data both correspond to true events.
15. The true event identification method according to claim 14, characterized in that: Based on the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data, judging whether the two sets of detection data both correspond to true events includes: Based on the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data, it is determined whether the two sets of detection data correspond to true coincidence events, and then it is determined whether the two sets of detection data both correspond to true events: If the two sets of detection data correspond to true coincident events, then the two sets of detection data correspond to true events; If the two sets of detection data do not correspond to a true coincidence event, then at least one set of detection data among the two sets of detection data does not correspond to a true event.
16. The true event identification method according to claim 15, characterized in that: Based on the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response line corresponding to the two sets of detection data, it is judged whether the two sets of detection data correspond to true coincidence events, including: According to whether the generatrix of the Compton cone corresponding to the two sets of detection data is collinear with the response line corresponding to the two sets of detection data, it is judged whether the two sets of detection data correspond to true coincidence events: If the generatrix of the Compton cone corresponding to the two sets of detection data is collinear with the response line corresponding to the two sets of detection data, it is determined that the two sets of detection data correspond to true coincidence events; Otherwise, it is determined that the two sets of detection data do not correspond to a true coincidence event.
17. The true event identification method according to claim 16, characterized in that: Whether the generatrix of the Compton cone corresponding to the two sets of detection data is collinear with the response line corresponding to the two sets of detection data includes: According to the relationship between the angles between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data, it is determined whether the generatrix of the Compton cone corresponding to the two sets of detection data is collinear with the response lines corresponding to the two sets of detection data.
18. The true event identification method according to claim 17, characterized in that: According to the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data, it is identified whether the generatrix of the Compton cone corresponding to the two sets of detection data is collinear with the response lines corresponding to the two sets of detection data, including: An angle threshold is set. If the angle between the busbar of the Compton cone and the response lines corresponding to the two sets of detection data exceeds the angle threshold, it is determined that the busbar of the Compton cone is not collinear with the response lines corresponding to the two sets of detection data. Otherwise, it is determined that the busbar of the Compton cone is collinear with the response lines corresponding to the two sets of detection data.
19. The true event identification method according to claim 18, characterized in that: Based on the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data, before judging whether the two sets of detection data both correspond to true events, the method further includes: The two groups of detection data that are selected to meet the time window are analyzed to determine whether each group of detection data corresponds to the Compton cone.
20. The true event identification method according to claim 19, characterized in that: The two groups of detection data that are selected and meet the time window are analyzed to determine whether each group of detection data corresponds to the Compton cone, including: Whether each group of detection data corresponds to a Compton cone is determined based on the number of detection data subsets corresponding to each of the two groups of detection data; the detection data subset is a data set corresponding to the CH address code of a scintillation crystal in a detection module.
21. The true event identification method according to claim 20, characterized in that: Judging whether each set of detection data corresponds to a Compton cone based on the number of detection data subsets corresponding to each of the two sets of detection data includes: If one set of detection data includes more than two detection data subsets, then the set of detection data corresponds to a Compton cone; If one set of detection data includes only one subset of detection data, then the set of detection data does not correspond to a Compton cone.
22. The true event identification method according to claim 18, characterized in that: Based on the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data, before judging whether the two sets of detection data both correspond to true events, the method further includes: If at least one of the energy sums corresponding to the two sets of detection data is not within the energy window: If both sets of detection data do not correspond to the Compton cone, it is impossible to determine whether the two sets of detection data correspond to true events, and the two sets of detection data are discarded; If at least one of the two sets of detection data corresponds to a Compton cone, it is determined whether the two sets of detection data both correspond to true events based on the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data.
23. The true event identification method according to claim 22, characterized in that: If at least one of the two sets of detection data corresponds to a Compton cone, then based on the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data, it is determined whether the two sets of detection data both correspond to true events, including: Based on the relationship between the generatrix of the Compton cone corresponding to the two sets of detection data and the response lines corresponding to the two sets of detection data, it is determined whether the two sets of detection data correspond to true coincidence events, and whether the two sets of detection data both correspond to true events: If only one of the two sets of detection data corresponds to a Compton cone, then based on the relationship between the generatrix of the Compton cone and the response lines corresponding to the two sets of detection data, by judging whether the two sets of detection data correspond to true coincidence events, it is judged whether the two sets of detection data both correspond to true events; If both sets of detection data correspond to Compton cones, then based on the relationship between the generatrix of the two Compton cones and the response lines corresponding to the two sets of detection data, it is determined whether the two sets of detection data correspond to true events by judging whether the two sets of detection data correspond to true events.
24. The true event identification method according to claim 23, characterized in that: Based on the relationship between the generatrix of the Compton cone and the response lines corresponding to the two sets of detection data, by judging whether the two sets of detection data correspond to true coincidence events, judging whether the two sets of detection data both correspond to true events includes: Determine the Compton cone corresponding to the set of detection data; According to whether the generatrix of the Compton cone is collinear with the response lines corresponding to the two sets of detection data, it is determined whether the two sets of detection data correspond to true coincidence events, and then it is determined whether the two sets of detection data both correspond to true events.
25. The true event identification method according to claim 24, characterized in that: According to whether the generatrix of the Compton cone is collinear with the response lines corresponding to the two sets of detection data, judging whether the two sets of detection data correspond to true coincidence events, and then judging that the two sets of detection data both correspond to true events, including: If the generatrix of the Compton cone is collinear with the response lines corresponding to the two sets of detection data, it is determined that the two sets of detection data correspond to true coincidence events, and accordingly, the two sets of detection data both correspond to true events; On the contrary, it is determined that the two groups of detection data do not correspond to a true coincidence event, and accordingly at least one group of detection data in the two groups of detection data does not correspond to a true event.
26. The true event identification method according to claim 25, characterized in that: Whether the generatrix of the Compton cone is colinear with the response lines corresponding to the two sets of detection data is determined based on the angle between the generatrix of the Compton cone and the response lines corresponding to the two sets of detection data.
27. The true event identification method according to claim 23, characterized in that: Based on the relationship between the generatrix of the two Compton cones and the response lines corresponding to the two sets of detection data, by judging whether the two sets of detection data correspond to true coincidence events, judging whether the two sets of detection data both correspond to true events includes: Obtain the Compton cones corresponding to the two sets of detection data respectively; According to whether the generatrixes of the two Compton cones are both collinear with the response lines corresponding to the two sets of detection data, it is determined whether the two sets of detection data correspond to true coincidence events, and then it is determined whether the two sets of detection data both correspond to true events.
28. The true event identification method according to claim 27, characterized in that: According to whether the generatrixes of the two Compton cones are both collinear with the response lines corresponding to the two sets of detection data, judging whether the two sets of detection data correspond to true coincidence events, and then judging whether the two sets of detection data both correspond to true events, including: If the generatrixes of the two Compton cones are both collinear with the response lines corresponding to the two sets of detection data, it is determined that the two sets of detection data correspond to true coincidence events, and accordingly, the two sets of detection data correspond to true events; On the contrary, if at least one of the busbars of the two Compton cones is not colinear with the response lines corresponding to the two sets of detection data, it is determined that the two sets of detection data do not correspond to a true coincidence event, and accordingly, at least one of the two sets of detection data does not correspond to a true event.
29. The true event identification method according to claim 28, characterized in that: Whether the generatrixes of the two Compton cones are both collinear with the response lines corresponding to the two sets of detection data is determined based on the angles between the generatrixes of the two Compton cones and the response lines corresponding to the two sets of detection data.
30. The true event identification method according to claim 26 or 29, characterized in that: The method of determining whether the generatrixes of the two Compton cones are both collinear with the response lines corresponding to the two sets of detection data based on the angles between the generatrixes of the two Compton cones and the response lines corresponding to the two sets of detection data comprises: An angle threshold is set. If the angle between the busbar of the Compton cone and the response lines corresponding to the two sets of detection data does not exceed the angle threshold, it is determined that the busbars of the Compton cone are collinear with the response lines corresponding to the two sets of detection data. Conversely, if the angle between the busbar of at least one Compton cone and the response lines corresponding to the two sets of detection data exceeds the angle threshold, it is determined that the busbar of the Compton cone is not collinear with the response lines corresponding to the two sets of detection data.
31. The true event identification method according to claim 30, characterized in that: The angle threshold is determined according to prior information, and the angle threshold is between 8 degrees and 12 degrees.
32. The true event identification method according to any one of claims 1-12 and 13-30, characterized in that: The true event identification method also includes: determining the Compton cone: the Compton cone is determined based on the scattering angle of scattered photons that undergo inter-crystal scattering and the flight direction of the scattered photons, and the scattering angle of the scattered photons and the flight direction of the scattered photons are determined based on each group of detection data corresponding to the Compton cone; the Compton cone includes one or two Compton cones respectively obtained based on one or two groups of detection data corresponding to the Compton cone.
33. The true event identification method according to claim 32, characterized in that: The Compton cone is a cone with the flight direction of the scattered photons as the rotation axis and the scattering angle of the scattered photons as the rotation angle, and the mother line direction of the cone is the direction of the incident photons.
34. The true event identification method according to claim 33, characterized in that: The deposition energy corresponding to the scattered photons when they are scattered is obtained based on each group of detection data corresponding to the Compton cone; the scattering angle of the scattered photons is obtained based on the energy of the incident photons and the deposition energy corresponding to the scattered photons when they are scattered.
35. The true event identification method according to claim 34, characterized in that: The corresponding deposition energy of scattered photons when scattered photons are scattered is obtained based on each set of detection data corresponding to the Compton cone, including: Counting the sum of the deposited energies corresponding to each detection data subset included in each group of detection data; Each group of detection data is sorted according to time information, and the detection data subset ranked first in initial time is used as the detection data subset corresponding to when scattering occurs; from the deposition energies corresponding to the detection data subsets included in each group of detection data, the deposition energy corresponding to the detection data subset ranked first in initial time is selected as the deposition energy corresponding to when the scattered photon occurs. The detection data subset is a data set corresponding to the CH address code of a scintillation crystal in a detection module.
36. The true event identification method according to claim 35, characterized in that: The sum of the deposited energies corresponding to each detection data subset included in each set of detection data is counted, including: The pulse waveform corresponding to each detection data subset is restored, and the pulse waveform is integrated to calculate the deposition energy corresponding to each detection data subset.
37. The true event identification method according to any one of claims 34-36, characterized in that: The formula for obtaining the scattering angle is as follows: Where E γ is the energy of the incident photon, E s is the deposited energy corresponding to the scattered photon when it is scattered, m e is the mass of the electron, c is the speed of light, and θ is the scattering angle.
38. The true event identification method according to claim 32, characterized in that: The flight direction of the scattered photon is determined based on the first energy deposition position and the second energy deposition position of the scattered photon corresponding to the occurrence of inter-crystal scattering. The direction of the line connecting the first energy deposition position of the scattered photon and the second energy deposition position of the scattered photon is the flight direction of the scattered photon.
39. The true event identification method according to claim 38, characterized in that: Each group of detection data is sorted according to time information, and the detection data subset ranked first in the initial time is used as the first detection data subset corresponding to the first energy deposition, and the energy deposition position obtained based on the first detection data subset is the first energy deposition position of the scattered photon; The detection data ranked second in the initial time is used as the second detection data subset corresponding to the second energy deposition, and the energy deposition position obtained based on the second detection data subset is the second energy deposition position of the scattered photon; The detection data subset is a data set corresponding to the CH address encoding of a scintillation crystal in a detection module.
40. The true event identification method according to claim 39, characterized in that: The energy deposition position is obtained based on the position information included in each group of detection data corresponding to the Compton cone and the reaction depth information corresponding to each group of detection data.
41. The true event identification method according to claim 40, characterized in that: The reaction depth is the depth at which scattered photons are deposited in the scintillation crystal; the energy deposition position is obtained based on the position information included in each group of detection data corresponding to the Compton cone and the reaction depth information corresponding to each group of detection data corresponding to the Compton cone, including: Based on the IP address code of the detection module corresponding to each group of detection data corresponding to the Compton cone and the CH address code of the scintillation crystal, the scintillation crystal where the scattered photons are deposited is determined, and the energy deposition position of the scattered photons on the scintillation crystal is determined based on the reaction depth information.
42. The true event identification method according to claim 41, characterized in that: The energy deposition position corresponding to the first energy deposition of the scattered photon is the energy deposition position of the scattered photon on the first scintillation crystal of the scintillation crystal array of the detection module obtained based on the first detection data subset and the corresponding reaction depth information; The energy deposition position corresponding to the second energy deposition of the scattered photon is the energy deposition position of the scattered photon on the second scintillation crystal of the scintillation crystal array of the same detection module obtained based on the second detection data subset and the corresponding reaction depth information; The energy deposition position is the center position of the depth section corresponding to the energy deposition depth on the scintillation crystal; the direction of the line connecting the first energy deposition position and the second energy deposition position on the scintillation crystal is the flight direction of the scattered photons.
43. The true event identification method according to any one of claims 40-42, characterized in that: The reaction depth is obtained based on a dual-end readout DOI detector or a single-end readout DOI detector.
44. An image reconstruction method, characterized in that: The image reconstruction method comprises: Acquire a true event by using the true event identification method described in any one of claims 1 to 43; An image is reconstructed based on the true event.
45. A true event identification device, characterized in that: The true event identification device comprises: A data acquisition module, configured to acquire detection data, wherein the detection data includes location information and time information; A time window screening module is configured to set a time window and an energy window, and screen out detection data within the time window based on the relationship between the energy sums corresponding to the two groups of detection data and the upper energy threshold of the energy window, wherein only two groups of detection data are included in the time window and the energy sums of the two groups of detection data do not exceed the upper energy threshold; The true event discrimination module is configured to judge whether each set of detection data corresponds to a true event based on the energy window and whether the two sets of detection data correspond to Compton cones.
46. An image reconstruction system, characterized in that: The image reconstruction system comprises: A true event acquisition module, configured to acquire true events using the true event identification method described in any one of claims 1 to 43; A reconstruction module is configured to perform image reconstruction based on the true event.
47. A digitizing 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 true event identification method according to any one of claims 1 to 43.
48. A storage medium, characterized in that The storage medium stores a computer program, which, when executed by a processor, implements the steps of the true event identification method as described in any one of claims 1 to 43.