Motion frame determination based on image data

Abstract

A system and method includes associating imaging event data with one of a plurality of bins based on a time associated with the imaging event data, determining that a time period of a first bin and a time period of a second bin are adjacent in time, determining whether a spatial characteristic of the imaging event data of the first bin is within a predetermined threshold of a spatial characteristic of the imaging event data of the second bin, and based on the determination, reconstructing one or more images based on the imaging event data of the first bin and the second bin.

Description

Background Technology

[0001] In conventional nuclear imaging, radiopharmaceuticals are introduced into the patient's body through injection or ingestion. The radiopharmaceutical emits gamma rays (in the case of single-photon emission computed tomography (SPECT) imaging) or emits positrons that annihilate with electrons to produce gamma rays (in the case of positron emission tomography (PET) imaging). An external detector system detects the emitted gamma rays and reconstructs an image based on this information.

[0002] The detection of emitted gamma rays occurs over a period of time during which the body may move unintentionally or due to natural physiological processes such as breathing and heartbeat. This movement may result in blurred images, particularly in the areas of the head, abdomen, chest, and heart.

[0003] Some systems address this issue by limiting the imaging volume. Other systems use external hardware (such as a 3D camera) to detect patient movement. This movement can be recorded as motion vectors that appear at specific times during data acquisition. The motion vectors are then used to determine periods of no movement or relatively low movement (i.e., motion frames), and the data acquired between motion frames is corrected to account for inter-frame motion.

[0004] The image can be reconstructed based on the corrected data. Data acquired during relatively significant motion is discarded and not used for image reconstruction. This data loss can adversely affect the quality of the reconstructed image.

[0005] The system is expected to identify motion frames solely based on imaging data, and compared to other motion correction systems, it retains a larger portion of the acquired data for image reconstruction. Attached Figure Description

[0006] Figure 1 This is a block diagram of a PET / CT imaging system according to some embodiments; Figure 2 The illustration shows event data collected over time according to some embodiments; Figure 3 Including flowcharts of the process of defining motion frames based on event data according to some embodiments; Figure 4 The illustration shows time-based binning of event data according to some embodiments. Figure 5 The illustration shows the determination of spatial information for each time-based bin of event data according to some embodiments; Figure 6 The illustration shows the merging of two time-based bins of event data according to some embodiments; Figure 7 The illustration shows the merging of a third time-based bin for event data according to some embodiments; Figure 8 The illustration shows the merging of two time-based bins of event data according to some embodiments; Figure 9 The illustration shows a cluster of event data and corresponding spatial information according to some embodiments. Figure 10 The illustration depicts a system for clustering, motion correction, and reconstruction of event data according to some embodiments; and Figure 11 The illustration shows a system for clustering, motion correction, and reconstruction of event data according to some embodiments. Detailed Implementation

[0007] The following description is provided to enable any person skilled in the art to make and use the described embodiments, and illustrates the best mode contemplated for implementing the described embodiments. However, various modifications will still be apparent to those skilled in the art.

[0008] Generally, some embodiments identify clusters of event data acquired during periods of no or low subject motion. Therefore, event data from a given cluster can be used to reconstruct an image with minimal motion artifacts.

[0009] Each cluster is associated with spatial information and an acquisition time window determined based on event data within that cluster. Motion between temporally adjacent clusters can be determined based on the spatial information of temporally adjacent clusters. The determined motion can be used for subsequent motion correction of the clustered event data before or during image reconstruction.

[0010] Advantageously, clusters can be determined without additional monitoring data provided by motion detection systems (such as external 3D cameras). Some embodiments also provide clustering of temporally adjacent event data, which contrasts with data clustering techniques that fail to consider temporal correlations between data (e.g., k-means clustering).

[0011] Figure 1 The illustration shows a PET / CT system 100 for performing one or more of the processes described herein. The embodiments are not limited to system 100. In particular, although the acquired data described herein is referred to as event data (i.e., corresponding to gamma-ray detection), the embodiments are not limited to SPECT and PET imaging systems. For example, the embodiments can be applied to data acquired via computed tomography (CT), magnetic resonance (MR), ultrasound, or other imaging modalities, as long as the acquired data is correlated with the acquisition time and three-dimensional location.

[0012] System 100 includes a gantry 110 defining a cavity 112. As is known in the art, gantry 110 houses a PET imaging assembly for acquiring PET image data and a CT imaging assembly for acquiring CT image data. The PET imaging assembly may include any number of gamma cameras employing any configuration known in the art. The CT imaging assembly may include one or more X-ray tubes and one or more corresponding X-ray detectors.

[0013] In conventional PET imaging, a tracer compound comprising a radionuclide is introduced into the patient's body via injection or ingestion. The radionuclide undergoes radioactive decay, producing a positron, which eventually encounters an electron and is annihilated. This annihilation produces two gamma photons traveling in roughly opposite directions. Therefore, an annihilation event is identified when two detectors positioned on opposite sides of the body detect the arrival of two gamma photons traveling in opposite directions within a specific coincidence time window.

[0014] Because the two gamma photons travel in roughly opposite directions, the positions of the two detectors determine the annihilation event along the response line (LOR) where it occurred. Time-of-flight (TOF) PET measures the difference between the detection times of the two gamma photons generated from the annihilation event. This difference can be used to estimate the specific location along the LOR where the annihilation event occurred. Therefore, each annihilation event can be represented by raw (i.e., list-mode) data specifying the three-dimensional location and time at which the event occurred.

[0015] The bed 115 and the base 116 are operable to allow a patient lying on the bed 115 to move into and out of the orifice 112. In some embodiments, the bed 115 is configured to translate on the base 116, and in other embodiments, the base 116 may move with the bed 115 or alternatively move from the bed 115.

[0016] The movement of the patient into and out of the cavity 112 allows for scanning of the patient using CT and PET imaging elements of the gantry 110. This scanning can be performed based on scanning parameters such as scan range and corresponding scan speed. According to some embodiments, during such a scan, the bed 115 and base 116 can provide continuous bed movement, as opposed to step-and-shoot motion.

[0017] The control system 120 may include any general-purpose or special-purpose computing system. Therefore, the control system 120 includes: one or more processing units 122 configured to execute processor-executable program code to cause the system 120 to operate as described herein; and a storage device 130 for storing the program code. The storage device 130 may include one or more fixed disks, solid-state random access memory, and / or removable media (e.g., thumb drives) mounted in a corresponding interface (e.g., a USB port).

[0018] Storage device 130 stores the program code of control program 131. One or more processing units 122 can execute control program 131 to control hardware components in conjunction with PET system interface 123, bed interface 125, and monitor interface 127 to move a patient into cavity 112, and during this movement, control a gamma camera to rotate around cavity 112 and detect coincidental events occurring within the body located in cavity 112. The detected events can be stored in memory 130 as PET data 134, which may include list pattern data and / or sine waves.

[0019] One or more processing units 122 may also execute control program 131 to coordinate with CT system interface 124 to cause radiation sources within gantry 110 to emit radiation toward the body within cavity 112 from different projection angles, and to control corresponding detectors to acquire two-dimensional CT data. CT data may be acquired substantially simultaneously with PET data as described above, and may be stored as CT data 136.

[0020] The storage device 130 also includes a motion correction procedure 132 for correcting the acquired PET data based on motion information. For example, embodiments may determine a motion vector representing patient movement between a first time period and a second time period. The motion correction procedure 132 may be executed to correct the PET data acquired during the second time period based on this motion vector, such that the corrected data is registered with the PET data acquired during the first time period.

[0021] Cluster definition 135 specifies the temporal range and spatial information corresponding to each of several clusters of event data. The temporal range of a cluster indicates the event data associated with that cluster (i.e., the event data acquired during that time range). The spatial information corresponding to each cluster can be used to determine motion vectors between clusters. The generation of cluster definition 135 according to some embodiments will be described in detail below.

[0022] A reconstruction procedure 133 can be executed to reconstruct PET images from PET data 134 using any known or becoming known reconstruction algorithm. As described below, two or more PET images can be reconstructed from corresponding clusters of event data acquired during periods of no or low motion, and / or a single PET image can be reconstructed from event data of all clusters that have been corrected based on one or more inter-cluster motion vectors. Such PET images can be stored in PET image 136.

[0023] PET and CT images can be transmitted to terminal 140 via terminal interface 126. Terminal 140 may include a display device and an input device coupled to system 120. Terminal 140 may display PET images, CT images, cluster definitions, and / or any other suitable images or data. Terminal 140 may receive user input for controlling the display of data, operation of system 100, and / or the processing described herein. In some embodiments, terminal 140 is a separate computing device, such as, but not limited to, a desktop computer, laptop computer, tablet computer, and smartphone.

[0024] Each component of system 100 may include other elements necessary for its operation, as well as additional elements for providing functionality beyond that described herein. Each functional component described herein may be implemented in computer hardware, in program code, and / or in one or more computing systems executing such program code known in the art. Such computing systems may include one or more processing units that execute processor-executable program code stored in a memory system.

[0025] For the purpose of describing some embodiments, Figure 2 This represents the acquired event data 200. Event data 200 describes events detected by a PET or SPECT scanner. Each event is described by its event time (i.e., the time when the event occurred) and its three-dimensional event location, the event time being determined by its location along an axis. t The location of this three-dimensional event is represented by its position. Figure 2 The event location can be represented using a three-dimensional graphic. The event location can be represented using any suitable coordinate system, such as, but not limited to, the coordinate system of the imaging system 100. As is known in the art, event data 200 can be acquired in a list mode.

[0026] Figure 3This includes a flowchart of process 300 for defining motion frames based on event data according to some embodiments. Flowchart 300 and other processes described herein can be executed using any suitable combination of hardware and software. The software program code embodying these processes can be stored by any non-transitory tangible medium, including fixed disks, volatile or non-volatile random access memory, DVDs, flash drives, and magnetic tapes. Embodiments are not limited to the examples described below.

[0027] Initially, multiple event data are acquired at S305. In some embodiments, the acquired event data may include list-pattern PET data as described above. The event data may be acquired by an imaging system separate from the system performing the remainder of process 300. The event data may initially be acquired in an imaging theatre, where process 300 is performed hours, days, months, etc., after the acquisition. Furthermore, although the acquired data is described as event data, any data associated with the acquisition time and location may be acquired at S305.

[0028] Next, at S310, each event data is assigned to (i.e., associated with) one of multiple warehouses. Each of these warehouses is associated with a time period, and the data for an event is assigned to a specific warehouse associated with the time period that includes that event. Regarding Figure 2 Event data 200, Figure 4 The illustration shows warehouse assignment at S310 according to some embodiments.

[0029] Figure 4 Eight bins, T0-T7, are shown, each containing a corresponding portion of data 200. Bin T0 includes all event data collected between times t0 and t1, bin T1 includes all event data collected between times t1 and t2, and so on. In some embodiments, each bin represents a time period of 1 second, and hundreds of bins can be used. The embodiments are not limited to any bin duration or number of bins.

[0030] At S315, spatial information (SI) associated with each warehouse is determined. This determination is based on the location associated with the event data assigned to each warehouse. In a non-exhaustive example, S315 includes determining the spatial location associated with warehouse T0 based on the location associated with each event data assigned to warehouse T0. At this point, each event data of warehouse T0 is associated with a three-dimensional location. Therefore, the spatial location determined at S315 is a three-dimensional location around which the location of each event data of warehouse T0 is uniformly distributed. The determination at S315 can employ any suitable algorithm known or becoming known.

[0031] Figure 5This refers to bins T0-T7 after the corresponding SI has been determined, according to some embodiments. As shown, bin T0 is associated with SI0, bin T1 with SI1, and so on. S315 may include: determining one or more of any spatial characteristics associated with each bin. This characteristic may be a representative location other than the geometric distribution center. As will be apparent from the description below, the SI determined for each bin at S315 may include a value / vector / equation representing a spatial entity that allows for the identification and quantification of movement between sets of event data.

[0032] At S320, the merging of warehouses into clusters begins. S320 includes determining whether the SI of the "next" warehouse is within a predetermined threshold of the SI of the "current" warehouse. On the first execution, the determination at S320 assumes that the first-in-time warehouse (i.e., warehouse T0) is the "current" warehouse.

[0033] In some embodiments, the predetermined threshold may include a scalar value. Therefore, the determination at S320 may include: determining the distance between the three-dimensional coordinates (or lines) of SI0 and the three-dimensional coordinates (or lines) of SI1, and determining whether the distance is less than the scalar value. Depending on the nature of the determined SI, any other measure of spatial distance may be used at S320. If the SI of a warehouse is represented by a curve or other multi-point entity, then S320 may include any suitable determination of the distance between two such entities.

[0034] If the determination at S320 is affirmative, the process proceeds to S325. At S325, the current warehouse and the next warehouse are merged into a combined warehouse. Figure 6 The illustration depicts the merging of warehouses T0 and T1 according to some embodiments. The merged (i.e., composite) warehouse is labeled T0T1. The merging of warehouses at S325 consists of the following operation: assigning all event data of the original warehouses T0 and T1 to the merged warehouse T0T1. As illustrated, the time period associated with the merged warehouse T0T1 is t0-t2. Therefore, the event data of the merged warehouse T0T1 includes all event data associated with the time between t0 and t2.

[0035] The merged warehouse T0T1 is associated with SI0. This association is used to ensure that any other warehouse merged with warehouse T0T1 is associated with SI within a predetermined threshold of SI0.

[0036] The process then continues to S335 to determine if there is another warehouse following the "next" warehouse. In this example, since warehouse T2 follows warehouse T1 in time, the determination at S335 is affirmative, and the process returns to S320. At S320, SI2 of the next warehouse T2 is compared with SI0 of warehouse T0T1. It will be assumed that SI2 is within a predetermined threshold of SI0, and therefore at S325, warehouse T2 is merged with warehouse T0T1. Figure 7 The diagram illustrates this merging, where the merged warehouse T0T1T2 includes all event data from the original warehouses T0, T1, and T2, and is associated with SI0 and the time period t0-t3.

[0037] Since warehouse T3 follows warehouse T2 in time, the process proceeds from S325 to S335 and back to S320, as described above. It will be assumed that warehouse T3's SI3 is not within the predetermined threshold of SI0. Therefore, the process continues to S330 to assign the next warehouse (i.e., warehouse T3) as the current warehouse. Due to the existence of the next warehouse T4, the process then returns to S320 to determine whether the next warehouse T4's SI4 is within the predetermined threshold of the new "current" warehouse T3's SI3.

[0038] Process 300 continues as described above until it is determined at S335 that there is no "next" warehouse available for comparison with the current warehouse for evaluation. Figure 8 As shown, process 300 has been executed as follows: do not merge positions T3 and T4, do not merge positions T4 and T5, merge positions T5 and T6, and do not merge positions T5T6 and T7.

[0039] Once it is determined at S335 that there is no "next" bin available for evaluation against the current bin, it is determined at S340 whether the maximum number of iterations has been reached. The above description of S315 to S335 constitutes one iteration. Some embodiments may implement a predefined maximum number of such iterations. If it is determined that the maximum number of iterations has been reached, the process terminates.

[0040] If it is determined at S375 that the maximum number of iterations has not yet been reached, the process continues to S345. At S345, it is determined whether the repositories defined at the end of the most recently completed iteration are different from the repositories defined at the end of the immediately preceding iteration (i.e., whether the last iteration changed these repositories). If not, it is assumed that further iterations will not change these warehouses, and the process terminates.

[0041] If the determination at S345 is positive, the process returns to S315 to determine the SI associated with each currently defined warehouse (i.e., warehouses T0T1T2, T3, T4, T5T6, and T7). Figure 9The diagram illustrates this spatial information for each of bins T0T1T2, T3, T4, T5T6, and T7. The SI of bin T0T1T2... 012 It may differ from the previous SI0, due to SI 012 It is determined based on the event data of each of the previous warehouses T0, T1, and T2. Similarly, the SI of warehouses T5 and T6 is determined. 56 It may differ from the previous SI5, due to SI 56 It was determined based on the event data of each of the previous warehouses T5 and T5.

[0042] The process proceeds as described above to S320 to determine whether the SI of the next-in-time warehouse (e.g., SI3 of warehouse T3) is within the SI of the current warehouse (e.g., SI of cluster T0T1T2). 012 Within a predetermined threshold. In some embodiments, a different predetermined threshold is used for each iteration. Then, whenever the determination at S320 is positive, the process continues as described above to merge the bins.

[0043] As mentioned above, iteration continues until the maximum number of iterations has been performed, or the final iteration has not changed these bins. At the end of process 300, the existing bins can be referred to as clusters. Each cluster is associated with a time period, event data of events occurring during that time period, and spatial information determined based on that event data. Because the event data of each cluster occurs before all events of subsequent clusters in time, SI can be used to determine motion vectors between successive clusters. The motion vectors can then be used to perform motion correction on the event data before reconstruction. According to some embodiments, clusters associated with time periods shorter than a given threshold are ignored during subsequent processing because such clusters are assumed to be associated with periods of significant motion.

[0044] Figure 10 The illustrations depict clustering, motion correction, and reconstruction of event data according to some embodiments. Figure 10 List pattern data 1010 is shown, which associates each detected gamma-ray event with the detection time and other values ​​that enable the determination of the event's location. Clustering component 1020 receives list pattern data 1010 and operates as described above to define clusters of event data.

[0045] It will be assumed that five clusters—clusters AE—are identified. Each of clusters AE is associated with time period and spatial information, as described above. The time period of cluster A precedes that of cluster B, the time period of cluster B precedes that of cluster C, and so on. Each cluster includes a portion of list pattern data 1010, which corresponds to the events that occurred during the time period of that cluster.

[0046] List pattern data 1030 includes events from list pattern data 1010 that occurred during a time period associated with cluster A. Each of datasets 1040a-d includes list pattern data for the corresponding cluster and a motion vector relative to cluster A. The motion vector for cluster B may include a vector between spatial information of cluster A and spatial information of cluster B. Similarly, the motion vector for cluster C may be determined based on the vector between spatial information of cluster A and spatial information of cluster B and the vector between spatial information of cluster B and spatial information of cluster C (e.g., the sum of these two vectors).

[0047] Motion correction component 1050 can apply motion correction known in the art to transform the list-pattern data of each of the datasets 1040a-d to a reference frame of list-pattern data 1030. The transformation can be based on a motion vector of each of the datasets 1040a-d, which can indicate relative motion with respect to a reference frame of list-pattern data 1030 as described above. Next, reconstruction component 1060 generates image 1070 based on list-pattern data 1030 and the motion-corrected datasets 1040a-d. As known in the art, this data can be subjected to various data processing algorithms (attenuation correction, denoising, etc.) prior to reconstruction.

[0048] According to some embodiments, motion correction is applied during reconstruction. (Reference) Figure 10 The reconstruction component 1060 can generate image 1070 based on list pattern data 1030 and non-motion-corrected datasets 1040a-d.

[0049] Figure 11 The illustration depicts clustering and reconstruction of event data according to some embodiments. Clustering component 1120 receives list pattern data 1110 and operates as described above to identify event data 1130a-e corresponding to the respective clusters.

[0050] The reconstruction component 1140 operates separately on each of the event data 1130a-e to generate images 1150a-e, where each image corresponds to a specific event data in the event data 1130a-e. Therefore, Figure 11 The illustration shows the generation of images corresponding to each cluster of event data, without considering the motion vectors between them.

[0051] Those skilled in the art will appreciate that various adaptations and modifications can be configured to the embodiments described above without departing from the claims. Therefore, it is to be understood that the claims can be practiced in ways other than those specifically described herein.

Claims

1. A system for clustering, motion correction, and reconstruction of event data, comprising: Imaging equipment, used for: Multiple imaging event data are acquired, each of which is associated with the three-dimensional location and acquisition time of the corresponding imaging event; Processing system, used for: Each imaging event data is assigned to one of a plurality of bins based on the time associated with the imaging event data, each of the plurality of bins being associated with a corresponding time period; For each bin, spatial information is determined based on the location of the imaging event data assigned to that bin; Identify the first pair of warehouses among the plurality of warehouses associated with adjacent time periods; Determine whether the spatial information of the later warehouse in the first identified warehouse pair is within a predetermined threshold of the spatial information of the earlier warehouse in the first identified warehouse pair; and In response to determining that the spatial information of the later warehouse in the first identified warehouse pair is within a predetermined threshold of the spatial information of the earlier warehouse in the first identified warehouse pair: The first pair of warehouses identified are merged into a first combined warehouse, which is associated with a first time period, which corresponds to a combination of adjacent time periods of the first pair of warehouses among the plurality of warehouses; as well as The first image is reconstructed based on imaging event data assigned to the first pair of bins among the plurality of bins.

2. The system according to claim 1, wherein the processing system is used for: After the first combined warehouse is formed, the warehouses that are time-concurrent with the first combined warehouse are processed sequentially to form the second combined warehouse, wherein the sequential processing includes: Determine whether the spatial information of the warehouse that immediately follows the first combined warehouse in time is within a predetermined threshold of the spatial information of the first combined warehouse; In response to determining that the spatial information of the warehouse that immediately follows the first combined warehouse in time is not within a predetermined threshold of the spatial information of the first combined warehouse, the warehouse that immediately follows the first combined warehouse in time is set as the current warehouse. Determine whether the spatial information of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the spatial information of the current warehouse; In response to the determination that the spatial information of the next warehouse immediately following the current warehouse in time is not within a predetermined threshold of the spatial information of the current warehouse, the next warehouse is taken as the current warehouse, and the determination of whether the spatial information of the next warehouse immediately following the current warehouse in time is within the predetermined threshold of the spatial information of the current warehouse is repeated. In response to the determination that the spatial information of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the spatial information of the current warehouse, the next warehouse and the current warehouse are merged as a second pair of warehouses among the plurality of warehouses to form a second combined warehouse. The second combined warehouse is associated with a second time period, which corresponds to a combination of adjacent time periods of the second pair of warehouses among the plurality of warehouses. as well as The second image is reconstructed based on imaging event data assigned to the second pair of bins among the plurality of bins.

3. The system according to claim 1, wherein the processing system is used for: After the first combined warehouse is formed, the warehouses that are time-concurrent with the first combined warehouse are processed sequentially to form the second combined warehouse, wherein the sequential processing includes: Determine whether the spatial information of the warehouse that immediately follows the first combined warehouse in time is within a predetermined threshold of the spatial information of the first combined warehouse; In response to determining that the spatial information of the warehouse that immediately follows the first combined warehouse in time is not within a predetermined threshold of the spatial information of the first combined warehouse, the warehouse that immediately follows the first combined warehouse in time is set as the current warehouse. Determine whether the spatial information of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the spatial information of the current warehouse; In response to the determination that the spatial information of the next warehouse immediately following the current warehouse in time is not within a predetermined threshold of the spatial information of the current warehouse, the next warehouse is taken as the current warehouse, and the determination of whether the spatial information of the next warehouse immediately following the current warehouse in time is within the predetermined threshold of the spatial information of the current warehouse is repeated. In response to the determination that the spatial information of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the spatial information of the current warehouse, the next warehouse and the current warehouse are merged as a second pair of warehouses among the plurality of warehouses to form a second combined warehouse. The second combined warehouse is associated with a second time period, which corresponds to a combination of adjacent time periods of the second pair of warehouses among the plurality of warehouses. The spatial information of the first combined warehouse is determined based on the location of the imaging event data assigned to the first pair of warehouses among the plurality of warehouses; The spatial information of the second combined warehouse is determined based on the location of imaging event data assigned to the second pair of warehouses among the plurality of warehouses; The motion vector is determined based on the spatial information of the first and second combined compartments; and Motion correction is performed on the imaging event data assigned to the second pair of bins among the plurality of bins based on the motion vector. The reconstruction of the first image includes reconstructing the first image based on imaging event data assigned to a first pair of bins in the plurality of bins and motion-corrected imaging event data assigned to a second pair of bins in the plurality of bins.

4. The system according to claim 1, wherein the processing system is used for: The identifier is associated with the third warehouse in a time period that is adjacent to and follows the time period of the first warehouse. Determine whether the spatial information of the third warehouse is within the predetermined threshold of the spatial information of the first combined warehouse; and In response to determining that the spatial information of the third warehouse is within a predetermined threshold of the spatial information of the first combined warehouse, the third warehouse is merged into the first combined warehouse to form an updated first combined warehouse. The reconstruction of the first image includes reconstructing the first image based on imaging event data assigned to a first pair of bins and imaging event data assigned to a third bin.

5. The system according to claim 4, wherein the processing system is used for: After forming the updated first combined warehouse, warehouses that are time-concurrently following the updated first combined warehouse are processed sequentially to form a second combined warehouse, wherein the sequential processing includes: Determine whether the spatial information of the first combined warehouse that is immediately updated in time among the plurality of warehouses is within a predetermined threshold of the spatial information of the updated first combined warehouse; In response to determining that the spatial information of the first combined warehouse that is immediately updated in time among the plurality of warehouses is not within a predetermined threshold of the spatial information of the updated first combined warehouse, the warehouse of the first combined warehouse that is immediately updated in time among the plurality of warehouses is set as the current warehouse; Determine whether the spatial information of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the spatial information of the current warehouse; In response to the determination that the spatial information of the next warehouse immediately following the current warehouse in time is not within a predetermined threshold of the spatial information of the current warehouse, the next warehouse is taken as the current warehouse, and the determination of whether the spatial information of the next warehouse immediately following the current warehouse in time is within the predetermined threshold of the spatial information of the current warehouse is repeated. In response to the determination that the spatial information of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the spatial information of the current warehouse, the next warehouse and the current warehouse are merged as a second pair of warehouses among the plurality of warehouses to form a second combined warehouse. The second combined warehouse is associated with a second time period, which corresponds to a combination of adjacent time periods of the second pair of warehouses among the plurality of warehouses. The spatial information of the first combined warehouse is determined based on the location of the imaging event data assigned to the first pair of warehouses and the imaging event data assigned to the third warehouse. The spatial information of the second combined warehouse is determined based on the location of imaging event data assigned to the second pair of warehouses among the plurality of warehouses; The motion vector is determined based on the updated spatial information of the first and second combined compartments; and Motion correction is performed on the imaging event data assigned to the second pair of bins among the plurality of bins based on the motion vector. The reconstruction of the first image includes reconstructing the first image based on imaging event data assigned to a first pair of bins in the plurality of bins, imaging event data assigned to a third bin, and motion-corrected imaging event data assigned to a second pair of bins in the plurality of bins.

6. The system of claim 1, wherein the imaging event data includes list-mode positron emission tomography (PET) data, and the first image is a PET image.

7. A method for clustering, motion correction, and reconstruction of event data, comprising: Multiple imaging event data are acquired, each of which is associated with the three-dimensional location and acquisition time of the corresponding imaging event; Each imaging event data is associated with one of a plurality of bins based on the time associated with the imaging event data, each of the plurality of bins being associated with a corresponding time period; For each bin, a representative location is determined based on the location of imaging event data associated with that bin; Identify the first pair of warehouses among the plurality of warehouses associated with adjacent time periods; Determine whether the representative location of the later warehouse in the first pair of identified warehouses is within a predetermined threshold of the representative location of the earlier warehouse in the first pair of identified warehouses; as well as In response to determining that the representative location of the later warehouse in the first identified warehouse pair is within a predetermined threshold of the representative location of the earlier warehouse in the first identified warehouse pair: The first pair of warehouses identified are merged into a first combined warehouse, which is associated with a first time period, which corresponds to a combination of adjacent time periods of the first pair of warehouses among the plurality of warehouses; as well as The first image is reconstructed based on imaging event data associated with a first pair of bins among the plurality of bins.

8. The method of claim 7, further comprising: After the first combined warehouse is formed, the warehouses that are time-concurrent with the first combined warehouse are processed sequentially to form the second combined warehouse, wherein the sequential processing includes: Determine whether the representative location of the warehouse that immediately follows the first combined warehouse in time is within a predetermined threshold of the representative location of the first combined warehouse; In response to determining that the representative position of the warehouse that immediately follows the first combined warehouse in time is not within a predetermined threshold of the representative position of the first combined warehouse, the warehouse that immediately follows the first combined warehouse in time is set as the current warehouse. Determine whether the representative location of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the representative location of the current warehouse; In response to the determination that the representative location of the next warehouse immediately following the current warehouse in time is not within a predetermined threshold of the representative location of the current warehouse, the next warehouse is taken as the current warehouse, and the determination of whether the representative location of the next warehouse immediately following the current warehouse in time is within the predetermined threshold of the representative location of the current warehouse is repeated. In response to determining that the representative location of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the representative location of the current warehouse, the next warehouse and the current warehouse are merged as a second pair of warehouses in the plurality of warehouses to form a second combined warehouse. The second combined warehouse is associated with a second time period, which corresponds to a combination of adjacent time periods of the second pair of warehouses in the plurality of warehouses; and The second image is reconstructed based on imaging event data associated with the second pair of bins in the plurality of bins.

9. The method of claim 7, further comprising: After the first combined warehouse is formed, the warehouses that are time-concurrent with the first combined warehouse are processed sequentially to form the second combined warehouse, wherein the sequential processing includes: Determine whether the representative location of the warehouse that immediately follows the first combined warehouse in time is within a predetermined threshold of the representative location of the first combined warehouse; In response to determining that the representative position of the warehouse that immediately follows the first combined warehouse in time is not within a predetermined threshold of the representative position of the first combined warehouse, the warehouse that immediately follows the first combined warehouse in time is set as the current warehouse. Determine whether the representative location of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the representative location of the current warehouse; In response to the determination that the representative location of the next warehouse immediately following the current warehouse in time is not within a predetermined threshold of the representative location of the current warehouse, the next warehouse is taken as the current warehouse, and the determination of whether the representative location of the next warehouse immediately following the current warehouse in time is within the predetermined threshold of the representative location of the current warehouse is repeated. In response to determining that the representative location of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the representative location of the current warehouse, the next warehouse and the current warehouse are merged as a second pair of warehouses among the plurality of warehouses to form a second combined warehouse. The second combined warehouse is associated with a second time period, which corresponds to a combination of adjacent time periods of the second pair of warehouses among the plurality of warehouses. The representative location of the first combined cell is determined based on the location of imaging event data associated with the first pair of cells in the plurality of cells; The representative location of the second combined bin is determined based on the location of imaging event data associated with the second pair of bins in the plurality of bins; The motion vector is determined based on the representative locations of the first and second combined warehouses; and Motion correction is performed on the imaging event data associated with the second pair of bins in the plurality of bins based on the motion vectors. The reconstruction of the first image includes reconstructing the first image based on imaging event data associated with a first pair of bins in the plurality of bins and motion-corrected imaging event data associated with a second pair of bins in the plurality of bins.

10. The method of claim 7, further comprising: The identifier is associated with the third warehouse in a time period that is adjacent to and follows the time period of the first warehouse. Determine whether the representative location of the third warehouse is within a predetermined threshold of the representative location of the first combined warehouse; as well as In response to determining that the representative location of the third warehouse is within a predetermined threshold of the representative location of the first combined warehouse, the third warehouse is merged into the first combined warehouse to form an updated first combined warehouse. The reconstruction of the first image includes reconstructing the first image based on imaging event data associated with a first pair of bins in the plurality of bins and imaging event data associated with a third bin.

11. The method of claim 10, further comprising: After forming the updated first combined warehouse, warehouses that are time-concurrently following the updated first combined warehouse are processed sequentially to form a second combined warehouse, wherein the sequential processing includes: Determine whether the representative location of the warehouse that immediately follows the updated first combination warehouse in time among the multiple warehouses is within a predetermined threshold of the representative location of the updated first combination warehouse; In response to determining that the representative location of the warehouse of the first combined warehouse that is immediately updated in time among the plurality of warehouses is not within a predetermined threshold of the representative location of the updated first combined warehouse, the warehouse of the first combined warehouse that is immediately updated in time among the plurality of warehouses is set as the current warehouse. Determine whether the representative location of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the representative location of the current warehouse; In response to the determination that the representative location of the next warehouse immediately following the current warehouse in time is not within a predetermined threshold of the representative location of the current warehouse, the next warehouse is taken as the current warehouse, and the determination of whether the representative location of the next warehouse immediately following the current warehouse in time is within the predetermined threshold of the representative location of the current warehouse is repeated. In response to determining that the representative location of the next warehouse immediately following the current warehouse in time is within a predetermined threshold of the representative location of the current warehouse, the next warehouse and the current warehouse are merged as a second pair of warehouses among the plurality of warehouses to form a second combined warehouse. The second combined warehouse is associated with a second time period, which corresponds to a combination of adjacent time periods of the second pair of warehouses among the plurality of warehouses. The representative location of the updated first combined cell is determined based on the location of imaging event data associated with the first pair of cells in the plurality of cells and the location of imaging event data associated with the third cell. The representative location of the second combined bin is determined based on the location of imaging event data associated with the second pair of bins in the plurality of bins; The motion vector is determined based on the updated representative positions of the first and second combined modules; and Motion correction is performed on the imaging event data associated with the second pair of bins in the plurality of bins based on the motion vectors. The reconstruction of the first image includes reconstructing the first image based on imaging event data associated with a first pair of bins in the plurality of bins, imaging event data associated with a third bin, and motion-corrected imaging event data associated with a second pair of bins in the plurality of bins.

12. The method of claim 7, wherein the imaging event data comprises list-mode positron emission tomography (PET) data, and the first image is a PET image.

Images