Methods and apparatus for automated time alignment processes within medical imaging systems

A two-stage time alignment process with wider and narrower time windows improves time estimation in TOF PET scanners, enhancing image quality by reducing noise and random coincidences.

WO2026235692A1PCT designated stage Publication Date: 2026-11-12SIEMENS MEDICAL SOLUTIONS USA INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/US2025/028334
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2025-05-08
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Conventional time-of-flight (TOF) PET scanners suffer from inaccuracies in time and time difference detection, leading to suboptimal image quality in nuclear imaging systems.

Method used

A method involving two-stage time alignment process using wider and narrower time windows to detect coincident events, generating and adjusting time offset values to improve time estimation accuracy, followed by storing these values for subsequent image reconstruction.

Benefits of technology

Enhances the quality and accuracy of nuclear imaging by reducing noise and random coincidences, resulting in higher-quality reconstructed images.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2025028334_12112026_PF_FP_ABST
    Figure US2025028334_12112026_PF_FP_ABST
Patent Text Reader

Abstract

Systems and methods for timing alignment in time-of-flight medical imaging systems are disclosed. In some examples, a processing system receives list mode data from a positron emission tomography (PET) image scanner. The system detects first coincident events within the list mode data using a first time window of a first duration. In addition, the system generates first time offset values based on the first plurality of coincident events. Further, the system adjusts the time of each of the events within the list mode data based on the first time offset values. The system further detects second coincident events within the adjusted list mode data using a second time window of a second duration. The system generates second time offset values based on the second plurality of coincident events. In addition, the system generates final time offset values based on the first time offset values and the second time offset values.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Docket No.: 2025P02105WO

[0002] METHODS AND APPARATUS FOR AUTOMATED TIME ALIGNMENT PROCESSES WITHIN MEDICAL IMAGING SYSTEMS FIELD

[0003] [1] Aspects of the present disclosure relate in general to medical diagnostic systems and, more particularly, to generating imaging data and reconstructing images in nuclear imaging systems.

[0004] BACKGROUND

[0005] [2] Nuclear imaging systems can employ various technologies to capture images. For example, some nuclear imaging systems employ positron emission tomography (PET) to capture images. PET is a modality of nuclear medicine imaging technique that produces tomographic images representing the distribution of positron emitting isotopes within a body. Typically, a PET scanner can capture imaging data in multiple directions, and a 3-dimensional PET image can be reconstructed from the various imaging captures. The detector elements of the PET scanner can simultaneously detect a pair of annihilation photons emitted from a positron emitting radiotracer that has been injected into a patient, where the LOR is the path between the two detector elements. For example, a time-of-flight (TOF) PET scanner can capture imaging data and can attempt to detect coincident events along a line of response (LOR).

[0006] [3] To detect coincident events, TOF PET scanners measure the time it takes for the photons to reach the detectors. Based on these times, the TOF PET scanners can estimate the location of events. At least some TOF PET scanners attempt to improve these location estimations by attempting to detect time differences between coincident events. The more accurate these time and time difference determinations are, the higher the quality and accuracy of the resulting images can be. Conventional TOF PET scanners, however, suffer from time and time difference detection inaccuracies. As such, there are opportunities to address deficiencies in nuclear imaging systems.

[0007] SUMMARY

[0008] [4] Systems and methods for time-of-flight (TOF) data time alignment and corresponding image reconstruction are disclosed.Docket No.: 2025P02105WO

[0009] [5] In some embodiments, a computer-implemented method includes receiving TOF data characterizing a location and time of each of a plurality of events. The method also includes detecting a first plurality of coincident events within the TOF data using a first time window of a first duration. In addition, the method includes generating first time offset values based on the first plurality of coincident events. Further, the method includes adjusting the time of each of the first plurality of coincident events within the TOF data based on the first time offset values. The method also includes detecting a second plurality of coincident events within the adjusted list mode data using a second time window of a second duration. The method further includes generating second time offset values based on the second plurality of coincident events. The method also includes generating final time offset values based on the first time offset values and the second time offset values.

[0010] Further, the method includes storing the final time offset values in a memory device.

[0011] [6] In some embodiments, a non-transitory computer readable medium stores instructions that, when executed by at least one processor, cause the at least one processor to perform operations. The operations include receiving TOF data characterizing a location and time of each of a plurality of events. The operations also include detecting a first plurality of coincident events within the TOF data using a first time window of a first duration. In addition, the operations include generating first time offset values based on the first plurality of coincident events. Further, the operations include adjusting the time of each of the first plurality of coincident events within the TOF data based on the first time offset values. The operations also include detecting a second plurality of coincident events within the adjusted list mode data using a second time window of a second duration. The operations further include generating second time offset values based on the second plurality of coincident events. The operations also include generating final time offset values based on the first time offset values and the second time offset values. Further, the operations include storing the final time offset values in a memory device.

[0012] [7] In some embodiments, an apparatus includes a memory storing instructions, and at least one processor communicatively coupled the memory. The at least one processor is configured to execute the instructions to perform operations. The operations include receiving TOF data characterizing a location andDocket No.: 2025P02105WG

[0013] time of each of a plurality of events. The operations also include detecting a first plurality of coincident events within the TOF data using a first time window of a first duration. In addition, the operations include generating first time offset values based on the first plurality of coincident events. Further, the operations include adjusting the time of each of the first plurality of coincident events within the TOF data based on the first time offset values. The operations also include detecting a second plurality of coincident events within the adjusted list mode data using a second time window of a second duration. The operations further include generating second time offset values based on the second plurality of coincident events. The operations also include generating final time offset values based on the first time offset values and the second time offset values. Further, the operations include storing the final time offset values in a memory device.

[0014] BRIEF DESCRIPTION OF THE DRAWINGS

[0015] [8] The following will be apparent from elements of the figures, which are provided for illustrative purposes and are not necessarily drawn to scale.

[0016] [9] FIGS. 1 A and 1 B illustrate various portions of a nuclear imaging system, in accordance with some embodiments.

[0017]

[0010] FIG. 2 illustrates a block diagram of an example processing device that can perform one or more of the operations described herein, in accordance with some embodiments.

[0018]

[0011] FIGS. 3A and 3B are flowcharts of an example time alignment method, in accordance with some embodiments.

[0019]

[0012] FIGS. 4A, 4B, and 4C illustrate signals within corresponding coincidence detection windows, in accordance with some embodiments.

[0020]

[0013] FIG. 5A is a flowchart of an example time alignment method, in accordance with some embodiments.

[0021]

[0014] FIG. 5B is a flowchart of an example method to detect coincident events, in accordance with some embodiments.Docket No.: 2025P02105WO

[0022] DETAILED DESCRIPTION

[0023]

[0015] This description of the exemplary embodiments is intended to be read in connection with the accompanying drawings, which are to be considered part of the entire written description. Independent of the grammatical term usage, individuals with male, female, or other gender identities are included within the term.

[0024]

[0016] The exemplary embodiments are described with respect to the claimed systems as well as with respect to the claimed methods. Furthermore, the exemplary embodiments are described with respect to methods and systems for image reconstruction, as well as with respect to methods and systems for training functions used for image reconstruction. Features, advantages, or alternative embodiments herein can be assigned to the other claimed objects and vice versa. For example, claims for the providing systems can be improved with features described or claimed in the context of the methods, and vice versa. In addition, the functional features of described or claimed methods are embodied by objective units of a providing system. Similarly, claims for methods and systems for training image reconstruction functions can be improved with features described or claimed in context of the methods and systems for image reconstruction, and vice versa.

[0025]

[0017] Various embodiments of the present disclosure can employ machine learning methods or processes to provide clinical information from nuclear imaging systems. For example, the embodiments can employ machine learning methods or processes to reconstruct images based on captured measurement data, and provide the reconstructed images for clinical diagnosis. In some embodiments, machine learning methods or processes are trained, to improve the reconstruction of images.

[0026]

[0018] Medical images reconstructed from time-of-light (TOF) data, such as TOF list mode data, tend to display less sensitivity to data inconsistencies, such as systematic errors in correction parameters, than other reconstruction methods. The quality of TOF image reconstruction, however, strongly depends on the correctness of the time information. For these reasons, a time alignment (TA) process, also referred to as a timing calibration process, is typically performed by TOF data medical imaging systems, such as positron emission tomography (PET), PET / computed tomography (CT), or PET / magnetic resonance (MR) imaging systems. For example, in some TOF scanners, a uniform activity cylindricalDocket No.: 2025P02105WO

[0027] phantom is placed at the center of the field-of-view (FOV) of the scanner, and then time offsets for each crystal along various line of responses (LORs) are estimated. In some TOF medical imaging systems, the uniform activity cylindrical phantom is scanned during continuous bed motion (CBM), where the bed is moved through the FOV of the scanner. During CBM, the time offsets for each crystal are estimated. Generated TOs are stored in the system and used during data rebinning for every LOR TOF information shift.

[0028]

[0019] In some TOF medical imaging systems, time offsets for each crystal are estimated by matching a TOF center of mass that is determined from applying a modelling process to the TOF data, and a TOF center of mass that is determined based on the measured TOF data.1One advantage of this process, at least when using a symmetrical blurring kernel (e.g., a Gaussian kernel), is that the center of mass is independent of TOF resolution. Moreover, the process can be generalized for CBM acquisitions. The estimation of the TOF center of mass for a given LOR, however, can be hindered by high random fraction due to, for example, delay subtraction.

[0029]

[0020] As described further herein, the random fraction can be decreased by narrowing the time window around a TOF signal. Because the location of the TOF signal is not known at the beginning of a TA procedure, TO estimations can first be determined using a relatively wider TOF window to detect coincident events. Using these initial TO estimations, the TOF signal can then be relocated in the TOF dimension. For example, TOF information identifies signal location (e.g., location of detecting crystals). The TOF information for each event can be stored as a list mode file in memory. To relocate a TOF signal, the list mode file can be rewritten to adjust the signal location for a coincident event based on the corresponding initial TO estimations for the two detectors involved in forming the coincident event. A relatively narrower time window can then be used to detect coincident events for subsequent TO estimations. Final TO estimations can then be computed based on the initial TO estimations and the subsequent TO estimations.

[0030] 1See, e.g., U.S. Patent No. 10, 852, 449, incorporated herein in its entirety.Docket No.: 2025P02105WO

[0031]

[0021] For example, as described further herein, during a time alignment process an image scanning system can scan an object(e.g., uniform activity cylindrical phantom), and can generate TOF data, such as list mode data, that characterizes a location and time of each of multiple detection events. Based on the list mode data, the image scanning system can detect initial coincident events using a first time window of a corresponding duration. For example, the image scanning system can determine that events that fall within the first time window are coincident events. The duration ( / .e. , width) of the first time window can be relatively longer than the duration of the second time window discussed below.

[0032]

[0022] In addition, the image scanning system can generate first time offset values based on the initial coincident events. The first time offset values characterize time offsets between crystals along corresponding LORs. For instance, in one example of how the image scanning system can generate the first time offset values, the image scanning system can generate non-TOF projection data, as well as normalization data (e.g., normalization array) and attenuation data (e.g., attenuation factors), from the list mode data. The image scanning system can then reconstruct an image from the non-TOF projection data, normalization data, and attenuation data. For instance, the reconstructed image is reconstructed from the non-TOF projection data, corrected for scatter efficiency based on the normalization data, and corrected for attenuation by the attenuation data. Based on the reconstructed image, the image scanning system can model non-TOF and TOF projections, as well as non-TOF and TOF scatter.

[0033]

[0023] Further still, the image scanning system can generate center-of-mass (COM) model data based on the non-TOF and TOF projections, as well as non-TOF and TOF scatter. The COM model data can be generated for each LOR. The image scanning system can also generate COM measurement data based on the initial coincident events detected using the first time window and, in some examples, the normalization data. The image scanning system can generate the COM measurement data for each LOR.

[0034]

[0024] The image scanning system can then generate the first time offset values (e.g., for each LOR) based on the COM model data and the COM measurement data. For instance, the image scanning system can compare theDocket No.: 2025P02105WO

[0035] center-of-masses characterized by each of the COM model data and COM measurement data, and generate the first time offset values based on the comparisons. Once generated, the image scanning system can relocate the corresponding TOF signal based on the first time offset values. For example, the image scanning system can adjust the time of each of the multiple detection events within the TOF data based on the first time offset values.

[0036]

[0025] Further, the image scanning system can re-detect the coincident events within the adjusted list mode data using a second time window of a duration that is less than the duration of the first time window described above. The second time offset values can be generated, for instance, similar to how the first time offset values were generated, as described above. For instance, the image scanning system can generate additional COM measurement data based on the re-detected coincident events detected using the second time window and, in some examples, the normalization data. The image scanning system can generate the additional COM measurement data for each LOR. In some examples, the duration of the second time window can be a predetermined percentage (e.g., 25%, 50%, 75%, 85%, et .) of the duration of the first time window. The image scanning system can compare the center-of-masses characterized by each of the COM model data and the additional COM measurement data, and generate the second time offset values based on the comparisons.

[0037]

[0026] The image scanning system can then generate final time offset values based on the first time offset values and the second time offset values. For instance, the image scanning system can add the first time offset values with the corresponding second time offset values (e.g., corresponding to the same set of crystals and LOR) to generate the final time offset values. The image scanning system can then store the final time offset values within a memory device (e.g., data repository). The image scanning system can then receive list mode data based on a scan of a patient, and can adjust the list mode data based on the final time offset values.

[0038]

[0027] The above time alignment (e.g., calibration) process can be occasionally executed, such as periodically, each time the image scanning system is powered up, or before a scan, for instance. When a subject is scanned, the imageDocket No.: 2025P02105WC

[0039] scanning system can obtain the final time offset values from the memory device, and apply them to the generated list mode data to account for residual time offsets between crystals. The adjusted list mode data can then be provided for image reconstruction, thereby resulting in higher quality and accurate images. In some examples, the above time alignment process is carried out during CBM.

[0040]

[0028] Because the first time window is relatively wider than the second time window, the first time window can contain relatively more random noise. The center-of-mass method, however, can require the estimation of center-of-mass using all “valid” range, since the location of the TOP signal is not yet known. Once relocated using the initial TO estimations, however, the TOF signals will be located near the center of the first time window zero. As such, the first time window can be narrowed to the second time window, thereby causing noise contribution to be reduced. As such, while random coincidence may be included in the first stage of the time alignment process ( / .e., when using the first time window), they can be truncated at the second state ( / .e., when using the relatively narrower second time window).

[0041]

[0029] FIG. 1 A illustrates portions of a nuclear imaging system 100 that can generate time offsets based on detecting coincident events. As illustrated, nuclear imaging system 100 includes an image scanner 102 and time alignment logic 119. The image scanner 102 includes detector elements 104, where each detector element 104 includes corresponding crystals 106 that can detect the annihilated photons. The image scanner 102 can be, for instance, a PET scanner that can detect annihilation photons emitted from a positron emitting radiotracer that has been injected into a patient. Further, the image scanner 102 can generate TOF data 111 (e.g., list mode data) characterizing locations and detection times of detected events, and can transmit the TOF data 111 to the time alignment logic 119.

[0042]

[0030] The time alignment logic 119 can carry out a time alignment (e.g., calibration) process (e.g., during CBM) to generate time offsets between LORs of the crystals 106 of the various detector elements 104. The time alignment logic 119 can be implemented in hardware, such as in one or more field-programmable gate arrays (FPGAs), one or more application-specific integrated circuits (ASICs), one or more state machines, one or more computing devices, digital circuitry, or any other suitable circuitry. In some examples, all or parts of the time alignment logic 119 canDocket No.: 2025P02105WO

[0043] be implemented in software as executable instructions such that, when executed by one or more processors, cause the one or more processors to perform respective functions as described herein. The instructions can be stored in a non-transitory, computer-readable storage medium, for instance.

[0044]

[0031] For example, FIG. 2 illustrates a processing device 200 that can be employed by the image reconstruction system 104. Processing device 200 can implement, for example, one or more of the functions of image reconstruction system 104 described herein.

[0045]

[0032] Processing device 200 can include one or more processors 201 , working memory 202, one or more input / output devices 203, instruction memory 207, a transceiver 204, one or more communication ports 209, and a display 206, all operatively coupled to one or more data buses 208. Data buses 208 allow for communication among the various devices. Data buses 208 can include wired, or wireless, communication channels.

[0046]

[0033] Processors 201 can include one or more distinct processors, each having one or more cores. Each of the distinct processors can have the same or different structure. Processors 201 can include one or more central processing units (CPUs), one or more graphics processing units (GPUs), application specific integrated circuits (ASICs), digital signal processors (DSPs), and the like.

[0047] Processors 201 can be configured to perform a certain function or operation by executing code, stored on instruction memory 207, embodying the function or operation. For example, processors 201 can be configured to perform one or more of any function, method, or operation disclosed herein by executing instructions stored in instruction memory 207.

[0048]

[0034] Instruction memory 207 can store instructions that can be accessed (e.g., read) and executed by processors 201. For example, instruction memory 207 can be a non-transitory, computer-readable storage medium such as a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), flash memory, a removable disk, CD-ROM, any non-volatile memory, or any other suitable memory. For example, instruction memory 207 can store instructions that, when executed by the one or more processors 201, cause one orDocket No.: 2025P02105WO

[0049] more processors 201 to perform one or more of the functions of image reconstruction system 104, such as one or more of the functions of any of the transformation determination engine 112, image registration engine 114, and image reconstruction engine 116, as described herein.

[0050]

[0035] Processors 201 can store data to, and read data from, working memory 202. For example, processors 201 can store a working set of instructions to working memory 202, such as instructions loaded from instruction memory 207. Processors 201 can also use working memory 202 to store dynamic data created during the operation of processing device 200. Working memory 202 can be a random access memory (RAM) such as a static random access memory (SRAM) or dynamic random access memory (DRAM), or any other suitable memory.

[0051]

[0036] Input / output devices 203 can include any suitable device that allows for data input or output. For example, input / output devices 203 can include one or more of a keyboard, a touchpad, a mouse, a stylus, a touchscreen, a physical button, a speaker, a microphone, or any other suitable input or output device.

[0052]

[0037] Communication port(s) 209 can include, for example, a serial port such as a universal asynchronous receiver / transmitter (UART) connection, a Universal Serial Bus (USB) connection, or any other suitable communication port or connection. In some examples, communication port(s) 207 allows for the programming of executable instructions in instruction memory 207. In some examples, communication port(s) 209 allow for the transfer (e.g., uploading or downloading) of data, such as configuration data to configure the durations of the various window sizes described further herein.

[0053]

[0038] Display 206 can display user interface 205. User interfaces 205 can enable user interaction with processing device 200. For example, user interface 205 can be a user interface for an application that allows for the viewing of the final image volumes 191 described further herein. In some examples, a user can interact with user interface 205 by engaging (e.g., touching with a finger or stylus) an input / output device 203. In some examples, display 206 can be a touchscreen, where user interface 205 is displayed on the touchscreen.Docket No.: 2025P02105WO

[0054]

[0039] T ransceiver 204 allows for communication with a network, such as a Wi-Fi® network, an Ethernet network, a cellular network, or any other suitable communication network. For example, if operating in a Wi-Fi® network, transceiver 204 is configured to allow communications with the Wi-Fi® network (e.g., and, in some examples, other devices on the Internet to which the Wi-Fi® network is connected). Processor(s) 201 is operable to receive data from, or send data to, a network via transceiver 204.

[0055]

[0040] Referring back to FIG. 1A, as illustrated, the time alignment logic 119 includes wide time window (TW) time offset (TO) determination engine 120, narrow TWTO determination engine 122, final TO determination engine 124, and memory device 126. The wide TWTO determination engine 120 can receive the TOF data 111, and can detect coincident events using a time window of a predetermined duration (e.g., 8 nanoseconds to 12 nanoseconds, such as 10 nanoseconds). For instance, the wide TWTO determination engine 120 can determine, based on the TOF data 111, that a first detector 104 detected an event within the time window, and that a second detector 104, located along a LOR with the first detector 104, also detected an event within the time window. The time window used by the wide TW TO determination engine 120 can be longer (e.g., wider) in duration that the window used by the narrow TWTO determination engine 122 to detect events, as described further below.

[0056]

[0041] FIG. 4A, for example, illustrates a graph 400 that includes a signal 401 (e.g., TOF signal) that is detected within a time window 403. The time window 403 can have a predetermined duration, such as a duration in the range of 4 to 12 nanoseconds or 4 to 10 nanoseconds. The signal 401 is illustrated as such for simplicity, but it is to be understood that the signal 401 can include random coincidences and noise. For example, FIG. 4C illustrates a signal 423 that includes a true coincident event (e.g., such as the true coincident event characterized by signal 401) as well as random coincidences and / or noise.

[0057]

[0042] Referring back to FIG. 1 A, the wide TW TO determination engine 120 can generate first time offset values 123 based on the detected coincident events. The wide TW TO determination engine 120 can generate the first time offset valuesDocket No.: 2025P02105WO

[0058] 123 using any suitable process, such as any of the center-of-mass based processes described herein (e.g., as described with respect to FIG. 3A below).

[0059]

[0043] As an example, the wide TW TO determination engine 120 can generate non-TOF projection data, as well as normalization data and attenuation data, from the TOF data 111. The wide TW TO determination engine 120 can then reconstruct an image from the non-TOF projection data, normalization data, and attenuation data. Based on the reconstructed image, the wide TWTO determination engine 120 can generate model data characterizing modelled non-TOF and TOF projections, as well as non-TOF and TOF scatter. Further still, the wide TW TO determination engine 120 can generate COM model data based on the non-TOF and TOF projections, as well as non-TOF and TOF scatter.

[0060]

[0044] The wide TW TO determination engine 120 can also generate COM measurement data based on the coincident events detected using the time window of the predetermined duration and, in some examples, the normalization data. The wide TW TO determination engine 120 can then generate the first time offset values 123 (e.g., for each LOR) based on the COM measurement data and COM measurement data. For instance, the wide TWTO determination engine 120 can compare the center-of-masses characterized by each of the COM measurement data and COM measurement data, and generate the first time offset values 123 based on the comparisons.

[0061]

[0045] The narrow TWTO determination engine 122 can receive the first time offset values 123 from the wide TW TO determination engine 120, as well as the TOF data 111 from the image scanner 102. The narrow TW TO determination engine 122 can generate adjusted TOF data by relocating the TOF signals characterized by the TOF data 111 based on the corresponding first time offset values 123. For example, the narrow TWTO determination engine 122 can adjust the time of each of the detection events within the TOF data 111 based on the first time offset values 123.

[0062]

[0046] Further, the narrow TWTO determination engine 122 can detect the coincident events within the adjusted TOF data using a time window of a duration that is less (e.g., narrower) than the duration of the time window used by the wideDocket No.: 2025P02105WO

[0063] TW TO determination engine 120 to detect the coincident events. As described herein, among other advantages, this second stage use of a narrow time window can reduce random fraction, as most of the “true” signal outside of the narrow window is likely zero and contaminated by “false” signal randoms events.

[0064]

[0047] FIG. 4B, for example, illustrates a graph 450 that includes a signal 411 that is detected within a time window 413. The signal 411 can be the signal 401 of FIG. 4A after being relocated based on the first time offset values 123, for instance. The time window 413 can have a predetermined duration, such as a predetermined percentage of the duration of the time window 403. For instance, the duration of the time window 413 can be in the range of 25% to 85% of the duration of the time window 403. In some examples, the time window 413 has a duration in the range of 1.5 to 10.2 nanoseconds.

[0065]

[0048] Referring back to FIG. 1A, the narrow TWTO determination engine 122 can generate second time offset values 125 based on the detected additional coincident events. The narrow TW TO determination engine 122 can generate the second time offset values 125 using any suitable process, such as any of the center-of-mass based processes described herein (e.g., as described with respect to FIG.

[0066] 3B below). For example, the narrow TW TO determination engine 122 can generate non-TOF projection data, as well as normalization data and attenuation data, from the TOF data 111. The narrow TW TO determination engine 122 can then reconstruct an image from the non-TOF projection data, normalization data, and attenuation data. Based on the reconstructed image, the narrow TW TO determination engine 122 can generate model data characterizing modelled non-TOF and TOF projections, as well as non-TOF and TOF scatter.

[0067]

[0049] Further still, the narrow TW TO determination engine 122 can generate COM model data based on the non-TOF and TOF projections, as well as non-TOF and TOF scatter. The narrow TW TO determination engine 122 can also generate COM measurement data based on the detected additional coincident events and, in some examples, the normalization data. The narrow TWTO determination engine 122 can then generate the second time offset values 125 (e.g., for each LOR) based on the COM measurement data and COM measurement data. For instance, the narrow TW TO determination engine 122 can compare the center-of-massesDocket No.: 2025P02105WO

[0068] characterized by each of the COM measurement data and COM measurement data, and generate the second time offset values 125 based on the comparisons.

[0069]

[0050] The final TO determination engine 124 can receive the first time offset values 123 from the wide TWTO determination engine 120 and the second time offset values 125 from the narrow TWTO determination engine 122. Further, the final TO determination engine 124 can generate final time offset values 127 based on the first time offset values 123 and the second time offset values 125. For instance, the final TO determination engine 124 can add the first time offset values 123 with the corresponding second time offset values 125 (e.g., corresponding to the same set of crystals and LOR) to generate the final time offset values 127. The final TO determination engine 124 can store the final time offset values 127 within the memory device 126.

[0070]

[0051] FIG. 1 B illustrates portions of a nuclear imaging system 100 that can reconstruct an image using generated time offsets, such as the final time offset values 127 described with respect to FIG. 1A. As illustrated, the nuclear imaging system 100 includes the image scanner 102 and image generation logic 160. The image generation logic 160 can be implemented in hardware, such as in one or more field-programmable gate arrays (FPGAs), one or more application-specific integrated circuits (ASICs), one or more state machines, one or more computing devices, digital circuitry, or any other suitable circuitry. In some examples, all or parts of the image generation logic 160 can be implemented in software as executable instructions such that, when executed by one or more processors, cause the one or more processors to perform respective functions as described herein. The instructions can be stored in a non-transitory, computer-readable storage medium, for instance.

[0071]

[0052] As illustrated, the image generation logic 160 includes coincidence determination engine 150, image reconstruction engine, and the memory device 126. The coincidence determination engine 150 can receive TOF data 111 (e.g., list mode data) from the image scanner 102. The TOF data 111 can be generated based on scans of a subject in the image scanner’s 102 field of view (FOV). The coincidence determination engine 150 can also receive the final time offset values 127 from the memory device 126. To detect coincident events, the coincidence determination engine 150 applies a window to the TOF data 111. The window (e.g.,Docket No.: 2025P02105WO

[0072] time window 413) can have a duration the same as the duration of the window applied by the narrow TWTO determination engine 122 (e.g., duration of time window 413). Further, for each detected event, the coincidence determination engine 150 can generate adjusted TOF data 151 by relocating the TOF data 111 (e.g., TOF signals) based on the corresponding final time offset values 127. For example, the coincidence determination engine 150 can adjust the time of each of the multiple detection events within the TOF data 111 based on the final time offset values 127 to generate adjusted TOF data 151.

[0073]

[0053] Image reconstruction engine 152 can receive the adjusted TOF data 151 from the coincidence determination engine 150 and, in some examples, can receive co-modality data 103 (e.g., p-map) from a co-modality scanner 162 (e.g., a CT or MR scanner). Further, the image reconstruction engine 125 applies an image reconstruction process to the adjusted TOF data 151 and, in some examples, to the co-modality data 103, to generate a reconstructed final image volume 191. For example, in applying the image reconstruction process, the image reconstruction engine 152 can correct the adjusted TOF data 151 for attenuation based on the co-modality data 103. In some instances, the image reconstruction engine 125 inputs the adjusted TOF data 151 to a trained machine learning model, such as a trained CNN, to generate the final image volume 191. In some instances, the image reconstruction engine 125 inputs the adjusted TOF data 151 and the co-modality data to a trained machine learning model, such as a trained CNN, to generate the final image volume 191.

[0074]

[0054] The image reconstruction engine 152 can store the final image volume 191 in the memory device 126. The memory device 126 can be any suitable memory device, such as a server, cloud-based storage, or a local memory device, for instance. Additionally or alternatively, the image reconstruction engine 152 can store the adjusted TOF data 151 in the memory device 126. In some examples, the image reconstruction engine 152 provides the final image volume 191 for display, such as to display 194.

[0075]

[0055] FIGS. 3A and 3B illustrate a time alignment procedure, such as one that can be performed by the image scanning system 100 described herein. TheDocket No.: 2025P02105WO

[0076] time alignment procedure can be carried out occasionally such as before each scan, and, in some examples, during CBM.

[0077]

[0056] With reference to FIG. 3A, process 300 begins at block 305 where list mode data is received. The list mode data can be generated by, and received from, the image scanner 102 based on scanning a patient, for example. At block 310, non-TOF projection data is generated based on applying a projection algorithm (e.g., model) to the list mode data and, at block 312, normalization data (e.g., normalization arrays) is generated based on applying a normalization algorithm (e.g., model) to the list mode data. In some examples, at block 307, a CT scan is captured (e.g., of the patient), and at block 314 attenuation data (e.g., attenuation coefficients) is generated based on applying an attenuation algorithm to the CT data.

[0078]

[0057] Further, at block 320, an image is reconstructed from the non-TOF projection data, normalization data, and, in some examples, the attenuation data. For instance, the reconstructed image is reconstructed from the non-TOF projection data, corrected for scatter efficiency based on the normalization data, and corrected for attenuation by the attenuation data. In some examples, the non-TOF projection data, normalization data, and attenuation data can be input to a reconstruction algorithm (e.g., trained CNN) that outputs the reconstructed image. Further, based on the reconstructed image, at block 330 non-TOF and TOF projections are modelled, and non-TOF and TOF scatter are modelled.

[0079]

[0058] At block 350, a TOF center-of-mass is modelled for each LOR based on the modelled non-TOF and TOF projections and non-TOF and TOF scatter. At block 340, a measured TOF center-of-mass is computed for each LOR based on the list mode data and, in some examples, the normalization data. To compute the measured TOF center-of-mass, coincident events of the list mode data are detected using a relatively wide time window (e.g., a time window with a longer duration than the time window described with respect to FIG. 3B). For example, the time window can be time window 403.

[0080]

[0059] Further, at block 360, first TOF time offsets are determined based on the modelled TOF center-of-mass generated at block 350, and the measured TOF center-of-mass generated at block 340. For example, the modelled TOF center-of-Docket No.: 2025P02105WO

[0081] mass and measured TOF center-of-mass can be compared, and the first TOF time offsets can be generated based on the comparison.

[0082]

[0060] Further, at block 370, the TOF signals, as characterized by the list mode data, can be relocated based on the first TOF time offsets. For example, the time of the detection events can be adjusted based on the first TOF time offsets.

[0083]

[0061] Referring now to FIG. 3B, process 380 begins at block 382 where additional list mode data is received. At block 384, non-TOF projection data is generated based on applying a projection algorithm (e.g., model) to the list mode data and, at block 392, normalization data (e.g., normalization arrays) is generated based on applying a normalization algorithm (e.g., model) to the list mode data. In some examples, at block 317, a CT scan is captured (e.g., of the patient), and at block 324 attenuation data (e.g., attenuation coefficients) is generated based on applying an attenuation algorithm to the CT data.

[0084]

[0062] Further, at block 386, an image is reconstructed from the non-TOF projection data, normalization data, and, in some examples, the attenuation data. Further, based on the reconstructed image, at block 388 non-TOF and TOF projections are modelled, and non-TOF and TOF scatter are modelled.

[0085]

[0063] At block 390, a TOF center-of-mass is modelled for each LOR based on the modelled non-TOF and TOF projections and non-TOF and TOF scatter. At block 394, a measured TOF center-of-mass is computed for each LOR based on the additional list mode data and, in some examples, the normalization data. To compute the measured TOF center-of-mass, coincident events of the additional list mode data are detected using a relatively narrow time window (e.g., a time window with a shorter duration than the time window described with respect to FIG. 3A). For example, the time window can be time window 413.

[0086]

[0064] Further, at block 396, second TOF time offsets are determined based on the modelled TOF center-of-mass generated at block 390, and the measured TOF center-of-mass generated at block 394. For example, the modelled TOF center-of-mass and measured TOF center-of-mass can be compared, and the second TOF time offsets can be generated based on the comparison.Docket No.: 2025P02105WO

[0087]

[0065] At block 398, final TOF time offsets are generated based on the first TOF time offsets (generated at block 360) and second TOF time offsets (generated at block 396). For example, each final TOF time offset can be computed by adding a first TOF time offset to its corresponding second TOF time offset. Further, and at block 399, the final TOF time offsets are stored in memory, such as in memory device 126.

[0088]

[0066] FIG. 5A is a flowchart of an example time alignment method 500. The method can be performed by, for example, the nuclear imaging system 100. The time alignment method 500 can be occasionally performed by the nuclear imaging system 100, such as part of occasional (e.g., periodical) timing calibration operations.

[0089]

[0067] Beginning at block 502, list mode data is received. The list mode data characterizes a location and time of each of a plurality of events. The time alignment logic 119 can receive the TOF data 111 from the image scanner 102, for example. At block 504, a first plurality of coincident events are detected within the list mode data using a first time window of a first duration. For example, the time alignment logic 119 can detect coincident events using the time window 403.

[0090]

[0068] Further, at block 506, first time offset values are generated based on the first plurality of coincident events. For instance, the time alignment logic 119 can generate first time offset values 123 based on any of the center-of-mass processes described herein (e.g., process 300). At block 508 the list mode data is relocated by adjusting the time of each of the plurality of events based on the first time offset values.

[0091]

[0069] Proceeding to block 510, a second plurality of coincident events are detected within the adjusted list mode data using a second time window of a second duration. For example, the time alignment logic 119 can detect the additional coincident events using the time window 413, for instance. At block 512, second time offset values are generated based on the second plurality of coincident events. For instance, the time alignment logic 119 can generate second time offset values 125 based on any of the center-of-mass processes described herein (e.g., process 380).Docket No.: 2025P02105WO

[0092]

[0070] At block 514 final time offset values are generated based on the first time offset values and the second time offset values. For instance, the time alignment logic 119 can add the first time offset values 123 with the second time offset values 125 to generate the final time offset values 127. Further, at block 516, the final time offset values are stored in a memory device. For example, the time alignment logic 119 can store the final time offset values 127 within the memory device 126.

[0093]

[0071] FIG. 5B is a flowchart of an example method 550 to detect coincident events. The method can be performed by, for example, the nuclear imaging system 100. The method 580 can be performed by the nuclear imaging system 100 during imaging, such as the imaging of a subject.

[0094]

[0072] Beginning at block 552, additional list mode data is received. The list mode data characterizes a location and time of each of a plurality of additional events. For example, image generation logic 160 can receive TOF data 111 from the image scanner 102. At block 554, the final time offset values are received from a memory device. For example, the image generation logic 160 can retrieve the final time offset values 127 from the memory device 126.

[0095]

[0073] At block 556, the time of each of the plurality of additional events are adjusted based on the final time offset values. For example, and as described herein, the time alignment logic 119 can adjust the times of the various events of the received TOF data 111 based on the final time offset values 127. For example, for a given event, the time alignment logic 119 can add a corresponding final time offset value 127 to the event’s time. Further, at block 558 at least one coincident event is detected withing the adjusted additional list mode data using the second time window of the second duration. For example, the time alignment logic 119 can determine that two events, after their times were adjusted, fall within the time window 413. As described herein, an image, such as the final image volume 191 , can be reconstructed based on the detected coincident events.

[0096]

[0074] The following is a list of non-limiting illustrative embodiments disclosed herein:Docket No.: 2025P02105WO

[0097]

[0075] Illustrative Embodiment 1 : A computer-implemented method comprising:

[0098] receiving time-of-flight (TOP) data characterizing a location and a time of each of a plurality of events;

[0099] detecting a first plurality of coincident events within the TOP data using a first time window of a first duration;

[0100] generating first time offset values based on the first plurality of coincident events;

[0101] adjusting the time of each of the first plurality of coincident events within the TOF data based on the first time offset values;

[0102] detecting a second plurality of coincident events within the adjusted TOF data using a second time window of a second duration;

[0103] generating second time offset values based on the second plurality of coincident events;

[0104] generating final time offset values based on the first time offset values and the second time offset values; and

[0105] storing the final time offset values in a memory device.

[0106]

[0076] Illustrative Embodiment 2: The computer-implemented method of illustrative embodiment 1, wherein the second duration of the second time window is less than the first duration of the first time window.

[0107]

[0077] Illustrative Embodiment 3: The computer-implemented method of any of illustrative embodiments 1-2, wherein detecting the first plurality of coincident events within the TOF data comprises detecting that the time of each of at least two events of the plurality of events fall within the same first time window.

[0108]

[0078] Illustrative Embodiment 4: The computer-implemented method of any of illustrative embodiments 1-3, comprising:

[0109] determining a measured center-of-mass based on the list mode data; and generating the first time offset values based on the measured center-of-mass.Docket No.: 2025P02105WO

[0110]

[0079] Illustrative Embodiment 5: The computer-implemented method of illustrative embodiment 4, comprising:

[0111] generating normalization data based on the list mode data; and determining the measured center-of-mass based on the normalization data.

[0112]

[0080] Illustrative Embodiment 6: The computer-implemented method of any of illustrative embodiments 4-5, comprising:

[0113] generating non-TOF projection data based on the list mode data; generating a reconstructed image based on the non-TOF projection data; generating non-TOF model projection data and TOF model projection data; generating non-TOF model scatter data and TOF model scatter data; determining a modelled center-of-mass based on the non-TOF model projection data, the TOF model projection data, the non-TOF model scatter data, and the TOF model scatter data; and

[0114] generating the first time offset values based on the modelled center-of-mass.

[0115]

[0081] Illustrative Embodiment 7: The computer-implemented method of illustrative embodiment 6, comprising generating the first time offset values based on a comparison of the measured center-of-mass and the modelled center-of-mass.

[0116]

[0082] Illustrative Embodiment 8: The computer-implemented method of any of illustrative embodiments 1-8, comprising:

[0117] receiving additional TOF data characterizing a location and a time of each of a plurality of additional events;

[0118] detecting a plurality of additional coincident events within the additional TOF data using the second time window of the second duration; and

[0119] adjusting the time of each of the plurality of additional coincident events within the additional TOF data based on the final time offset values.Docket No.: 2025P02105WO

[0120]

[0083] Illustrative Embodiment 9: The computer-implemented method of illustrative embodiment 8, comprising reconstructing an image based on the adjusted additional TOF data.

[0121]

[0084] Illustrative Embodiment 10: The computer-implemented method of any of illustrative embodiments 1-9, wherein generating the final time offset values comprises adding the first time offset values with the second time offset values.

[0122]

[0085] Illustrative Embodiment 11 : The computer-implemented method of any of illustrative embodiments 1-10, wherein the TOF data is list mode data.

[0123]

[0086] Illustrative Embodiment 12: A non-transitory, computer readable medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising:

[0124] receiving time-of-flight (TOF) data characterizing a location and a time of each of a plurality of events;

[0125] detecting a first plurality of coincident events within the TOF data using a first time window of a first duration;

[0126] generating first time offset values based on the first plurality of coincident events;

[0127] adjusting the time of each of the first plurality of coincident events within the TOF data based on the first time offset values;

[0128] detecting a second plurality of coincident events within the adjusted TOF data using a second time window of a second duration;

[0129] generating second time offset values based on the second plurality of coincident events;

[0130] generating final time offset values based on the first time offset values and the second time offset values; and

[0131] storing the final time offset values in a memory device.

[0132]

[0087] Illustrative Embodiment 13: The non-transitory, computer readable medium of claim 12, wherein the second duration of the second time window is less than the first duration of the first time window.Docket No.: 2025P02105WO

[0133]

[0088] Illustrative Embodiment 14: The non-transitory, computer readable medium of claim 12, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising detecting that the time of each of at least two events of the plurality of events fall within the same first time window.

[0134]

[0089] Illustrative Embodiment 15: The non-transitory, computer readable medium of claim 12, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising:

[0135] determining a measured center-of-mass based on the list mode data; and generating the first time offset values based on the measured center-of-mass.

[0136]

[0090] Illustrative Embodiment 16: The non-transitory, computer readable medium of illustrative embodiment 15, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising:

[0137] generating normalization data based on the list mode data; and determining the measured center-of-mass based on the normalization data.

[0138]

[0091] Illustrative Embodiment 17: The non-transitory, computer readable medium of any of illustrative embodiments 15-16, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising:

[0139] generating non-TOF projection data based on the list mode data; generating a reconstructed image based on the non-TOF projection data; generating non-TOF model projection data and TOF model projection data; generating non-TOF model scatter data and TOF model scatter data; determining a modelled center-of-mass based on the non-TOF model projection data, the TOF model projection data, the non-TOF model scatter data, and the TOF model scatter data; and

[0140] generating the first time offset values based on the modelled center-of-mass.Docket No.: 2025P02105WO

[0141]

[0092] Illustrative Embodiment 18: The non-transitory, computer readable medium of illustrative embodiment 17, comprising generating the first time offset values based on a comparison of the measured center-of-mass and the modelled center-of-mass.

[0142]

[0093] Illustrative Embodiment 19: The non-transitory, computer readable medium of any of illustrative embodiments 12-18, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising:

[0143] receiving additional TOP data characterizing a location and a time of each of a plurality of additional events;

[0144] detecting a plurality of additional coincident events within the additional TOF data using the second time window of the second duration;

[0145] adjusting the time of each of the plurality of additional coincident events within the additional TOF data based on the final time offset values; and

[0146] transmitting the adjusted additional TOF data for image reconstruction.

[0147]

[0094] Illustrative Embodiment 20: The non-transitory, computer readable medium of illustrative embodiment 19, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising reconstructing an image based on the adjusted additional TOF data.

[0148]

[0095] Illustrative Embodiment 21 : The non-transitory, computer readable medium of any of illustrative embodiments 12-20, where to generate the final time offset values, the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising adding the first time offset values with the second time offset values.

[0149]

[0096] Illustrative Embodiment 22: The non-transitory, computer readable medium of any of illustrative embodiments 12-21, wherein the TOF data is list mode data.Docket No.: 2025P02105WO

[0150]

[0097] Illustrative Embodiment 23: A system comprising:

[0151] a memory storing instructions; and

[0152] at least one processor communicatively coupled to the memory and configured to execute the instructions to:

[0153] receive time-of-flight (TOF) data characterizing a location and a time of each of a plurality of events;

[0154] detect a first plurality of coincident events within the TOF data using a first time window of a first duration;

[0155] generate first time offset values based on the first plurality of coincident events;

[0156] adjust the time of each of the first plurality of coincident events within the TOF data based on the first time offset values;

[0157] detect a second plurality of coincident events within the adjusted TOF data using a second time window of a second duration;

[0158] generate second time offset values based on the second plurality of coincident events;

[0159] generate final time offset values based on the first time offset values and the second time offset values; and

[0160] store the final time offset values in a memory device.

[0161]

[0098] Illustrative Embodiment 24: The system of illustrative embodiment 23, wherein the second duration of the second time window is less than the first duration of the first time window.

[0162]

[0099] Illustrative Embodiment 25: The system of any of illustrative embodiments 23-24, wherein the at least one processor is configured to execute the instructions to detect that the time of each of at least two events of the plurality of events fall within the same first time window.

[0163]

[0100] Illustrative Embodiment 26: The system of any of illustrative embodiments 23-25, wherein the at least one processor is configured to execute the instructions to:

[0164] determine a measured center-of-mass based on the list mode data; and generate the first time offset values based on the measured center-of-mass.Docket No.: 2025P02105WO

[0165]

[0101] Illustrative Embodiment 27: The system of illustrative embodiment 26, comprising:

[0166] generating normalization data based on the list mode data; and determining the measured center-of-mass based on the normalization data.

[0167]

[0102] Illustrative Embodiment 28: The system of any of illustrative embodiments 26-27, wherein the at least one processor is configured to execute the instructions to:

[0168] generate non-TOF projection data based on the list mode data; generate a reconstructed image based on the non-TOF projection data; generate non-TOF model projection data and TOF model projection data; generate non-TOF model scatter data and TOF model scatter data; determine a modelled center-of-mass based on the non-TOF model projection data, the TOF model projection data, the non-TOF model scatter data, and the TOF model scatter data; and

[0169] generate the first time offset values based on the modelled center-of-mass.

[0170]

[0103] Illustrative Embodiment 29: The system of illustrative embodiment 28, wherein the at least one processor is configured to execute the instructions to generate the first time offset values based on a comparison of the measured center-of-mass and the modelled center-of-mass.

[0171]

[0104] Illustrative Embodiment 30: The system of any of illustrative embodiments 23-29, wherein the at least one processor is configured to execute the instructions to:

[0172] receive additional TOF data characterizing a location and a time of each of a plurality of additional events;

[0173] detect a plurality of additional coincident events within the additional TOF data using the second time window of the second duration; and

[0174] adjust the time of each of the plurality of additional coincident events within the additional TOF data based on the final time offset values.Docket No.: 2025P02105WO

[0175]

[0105] Illustrative Embodiment 31: The system of illustrative embodiment 30, wherein the at least one processor is configured to execute the instructions to reconstruct an image based on the adjusted additional TOF data.

[0176]

[0106] Illustrative Embodiment 32: The system of any of illustrative embodiments 23-31 , wherein to generate the final time offset values, the at least one processor is configured to execute the instructions to add the first time offset values with the second time offset values.

[0177]

[0107] Illustrative Embodiment 33: The system of any of illustrative embodiments 23-32, wherein the TOF data is list mode data.

[0178]

[0108] Illustrative Embodiment 34: A system comprising:

[0179] a means for receiving time-of-flight (TOF) data characterizing a location and a time of each of a plurality of events;

[0180] a means for detecting a first plurality of coincident events within the TOF data using a first time window of a first duration;

[0181] a means for generating first time offset values based on the first plurality of coincident events;

[0182] a means for adjusting the time of each of the first plurality of coincident events within the TOF data based on the first time offset values;

[0183] a means for detecting a second plurality of coincident events within the adjusted TOF data using a second time window of a second duration;

[0184] a means for generating second time offset values based on the second plurality of coincident events;

[0185] a means for generating final time offset values based on the first time offset values and the second time offset values; and

[0186] a means for storing the final time offset values in a memory device.

[0187]

[0109] Illustrative Embodiment 35: The system of illustrative embodiment 34, wherein the second duration of the second time window is less than the first duration of the first time window.Docket No.: 2025P02105WO

[0188]

[0110] Illustrative Embodiment 36: The system of any of illustrative embodiments 34-35, comprising a means for detecting that the time of each of at least two events of the plurality of events fall within the same first time window.

[0189]

[0111] Illustrative Embodiment 37: The system of any of illustrative embodiments 34-36, comprising:

[0190] a means for determining a measured center-of-mass based on the list mode data; and

[0191] a means for generating the first time offset values based on the measured center-of-mass.

[0192]

[0112] Illustrative Embodiment 38: The system of illustrative embodiment 37, comprising:

[0193] a means for generating normalization data based on the list mode data; and a means for determining the measured center-of-mass based on the normalization data.

[0194]

[0113] Illustrative Embodiment 39: The system of any of illustrative embodiments 37-38, comprising:

[0195] a means for generating non-TOF projection data based on the list mode data; a means for generating a reconstructed image based on the non-TOF projection data;

[0196] a means for generating non-TOF model projection data and TOF model projection data;

[0197] a means for generating non-TOF model scatter data and TOF model scatter data;

[0198] a means for determining a modelled center-of-mass based on the non-TOF model projection data, the TOF model projection data, the non-TOF model scatter data, and the TOF model scatter data; and

[0199] a means for generating the first time offset values based on the modelled center-of-mass.Docket No.: 2025P02105WO

[0200]

[0114] Illustrative Embodiment 40: The system of illustrative embodiment 39, comprising a means for generating the first time offset values based on a comparison of the measured center-of-mass and the modelled center-of-mass.

[0201]

[0115] Illustrative Embodiment 41 : The system of any of illustrative embodiments 34-40, comprising:

[0202] a means for receiving additional TOF data characterizing a location and a time of each of a plurality of additional events;

[0203] a means for detecting a plurality of additional coincident events within the additional TOF data using the second time window of the second duration; and a means for adjusting the time of each of the plurality of additional coincident events within the additional TOF data based on the final time offset values.

[0204]

[0116] Illustrative Embodiment 42: The system of illustrative embodiment 41 , comprising a means for reconstructing an image based on the adjusted additional TOF data.

[0205]

[0117] Illustrative Embodiment 43: The system of any of illustrative embodiments 34-42, wherein the means for generating the final time offset values comprises adding the first time offset values with the second time offset values.

[0206]

[0118] Illustrative Embodiment 44: The system of any of illustrative embodiments 34-43, wherein the TOF data is list mode data.

[0207]

[0119] The apparatuses and processes are not limited to the specific embodiments described herein. In addition, components of each apparatus and each process can be practiced independent and separate from other components and processes described herein.

[0208]

[0120] The previous description of embodiments is provided to enable any person skilled in the art to practice the disclosure. The various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without the use of inventive faculty. The present disclosure is not intended to be limited to theDocket No.: 2025P02105WO

[0209] embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

Docket No.: 2025P02105WO What is claimed is:

1. A computer-implemented method comprising:receiving time-of-flight (TOF) data characterizing a location and a time of each of a plurality of events;detecting a first plurality of coincident events within the TOF data using a first time window of a first duration;generating first time offset values based on the first plurality of coincident events; adjusting the time of each of the first plurality of coincident events within the TOF data based on the first time offset values;detecting a second plurality of coincident events within the adjusted TOF data using a second time window of a second duration;generating second time offset values based on the second plurality of coincident events;generating final time offset values based on the first time offset values and the second time offset values; andstoring the final time offset values in a memory device.

2. The computer-implemented method of claim 1 , wherein the second duration of the second time window is less than the first duration of the first time window.

3. The computer-implemented method of claim 1 , wherein detecting the first plurality of coincident events within the TOF data comprises detecting that the time of each of at least two events of the plurality of events fall within the same first time window.Docket No.: 2025P02105WO4. The computer-implemented method of claim 1 , comprising:determining a measured center-of-mass based on the list mode data; and generating the first time offset values based on the measured center-of-mass.

5. The computer-implemented method of claim 4, comprising:generating normalization data based on the list mode data; anddetermining the measured center-of-mass based on the normalization data.

6. The computer-implemented method of claim 4, comprising:generating non-TOF projection data based on the list mode data; generating a reconstructed image based on the non-TOF projection data; generating non-TOF model projection data and TOF model projection data; generating non-TOF model scatter data and TOF model scatter data; determining a modelled center-of-mass based on the non-TOF model projection data, the TOF model projection data, the non-TOF model scatter data, and the TOF model scatter data; andgenerating the first time offset values based on the modelled center-of-mass.

7. The computer-implemented method of claim 6, comprising generating the first time offset values based on a comparison of the measured center-of-mass and the modelled center-of-mass.

8. The computer-implemented method of claim 1 , comprising:Docket No.: 2025P02105WO receiving additional TOF data characterizing a location and a time of each of a plurality of additional events;detecting a plurality of additional coincident events within the additional TOF data using the second time window of the second duration; andadjusting the time of each of the plurality of additional coincident events within the additional TOF data based on the final time offset values.

9. The computer-implemented method of claim 8, comprising reconstructing an image based on the adjusted additional TOF data.

10. The computer-implemented method of claim 1 , wherein generating the final time offset values comprises adding the first time offset values with the second time offset values.

11. The computer-implemented method of claim 1 , wherein the TOF data is list mode data.

12. A non-transitory, computer readable medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising:receiving time-of-flight (TOF) data characterizing a location and a time of each of a plurality of events;Docket No.: 2025P02105WO detecting a first plurality of coincident events within the TOF data using a first time window of a first duration;generating first time offset values based on the first plurality of coincident events; adjusting the time of each of the first plurality of coincident events within the TOF data based on the first time offset values;detecting a second plurality of coincident events within the adjusted TOF data using a second time window of a second duration;generating second time offset values based on the second plurality of coincident events;generating final time offset values based on the first time offset values and the second time offset values; andstoring the final time offset values in a memory device.

13. The non-transitory, computer readable medium of claim 12, wherein the second duration of the second time window is less than the first duration of the first time window.

14. The non-transitory, computer readable medium of claim 12, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising detecting that the time of each of at least two events of the plurality of events fall within the same first time window.Docket No.: 2025P02105WO 15. The non-transitory, computer readable medium of claim 12, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising:determining a measured center-of-mass based on the list mode data; and generating the first time offset values based on the measured center-of-mass.

16. The non-transitory, computer readable medium of claim 15, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations comprising:generating normalization data based on the list mode data; anddetermining the measured center-of-mass based on the normalization data.

17. A system comprising:a memory storing instructions; andat least one processor communicatively coupled to the memory and configured to execute the instructions to:receive time-of-flight (TOF) data characterizing a location and a time of each of a plurality of events;detect a first plurality of coincident events within the TOF data using a first time window of a first duration;generate first time offset values based on the first plurality of coincident events;Docket No.: 2025P02105WO adjust the time of each of the first plurality of coincident events within the TOF data based on the first time offset values;detect a second plurality of coincident events within the adjusted TOF data using a second time window of a second duration;generate second time offset values based on the second plurality of coincident events;generate final time offset values based on the first time offset values and the second time offset values; andstore the final time offset values in a memory device.

18. The system of claim 17, wherein the second duration of the second time window is less than the first duration of the first time window.

19. The system of claim 17, wherein the at least one processor is configured to execute the instructions to detect that the time of each of at least two events of the plurality of events fall within the same first time window.

20. The system of claim 17, wherein the at least one processor is configured to execute the instructions to:determine a measured center-of-mass based on the list mode data; and generate the first time offset values based on the measured center-of-mass.