PET imaging using multi-organ specific short ct scans

Abstract

A method for minimizing a patient's exposure to CT scan radiation during mu-map generation in a long-axis field-of-view (FOV) PET scan includes: performing a long-axis FOV PET scan on the patient; performing one or more truncated FOV CT scans on a region of interest in the patient's body; generating a truncated mu-map covering the truncated CT FOV; and generating a mu-map covering the full long-axis FOV of the PET scan by extending the truncated mu-map generated from the truncated FOV CT scan using the PET data to estimate missing mu-map data.

Description

Technical Field

[0001] This disclosure generally relates to a method for using one or more short CT scans whose field of view (FOV) may be shorter than that of a positron emission tomography (PET) scan, and provides PET reconstructed images with minimal artifacts. Furthermore, the truncated portion of the sampled CT scan helps determine whether the scattering obtained using the estimated mu-map and PET is correct, and can be used to improve the scattering scaling factor, scattering shape, and scattering tail fitting. Additionally, the estimated mu-map can be compared with the measured truncated mu-map to correct for any systematic errors in the measured data, such as timing offsets. Background Technology

[0002] CT scans and PET scans are well-known methods for diagnostic medical imaging. CT scans use multiple X-ray images acquired in multiple directions to generate three-dimensional images or multiple tomographic image "slices." PET scans use gamma-emitting radiopharmaceuticals ingested or injected into the patient. Multiple gamma-ray images are acquired in multiple directions to generate three-dimensional PET images or multiple slices. CT and PET scans provide different information. For example, CT scans typically have higher resolution and are superior in providing structural data on structures such as bones and organs. PET scans typically have lower resolution but provide more useful information about the functional condition of body tissues and systems, such as the cardiovascular system. For example, PET is superior in indicating the presence of soft tissue tumors or reduced blood flow to certain organs or regions of the body. The complementary intensity of CT and PET scans can be provided simultaneously by performing both methods in a single device and imaging step.

[0003] PET scanners with longer axial lengths are becoming increasingly popular for whole-body imaging. The longer axial length of the PET scanner reduces scan time for whole-body imaging and increases the overall sensitivity of the scanner. CT scans covering the same scan range as the PET field of view (FOV) are used to generate robust mu-maps for attenuation compensation. In this disclosure, the terms CT and mu-map are used interchangeably.

[0004] On the other hand, for organ-specific imaging, such as cardiac imaging, the axial length of a long PET scanner or multi-bed PET scan is typically longer than the region of the organ of interest in the patient's body, and will cover the organ of interest as well as other areas around it. Currently, when using a PET scanner with a long axial FOV, the entire (e.g., full) axial FOV of the scanner is scanned by CT to generate a mu-map. For example, in a brain scan using a PET scanner with an axial FOV of 26 cm, because the scanner FOV is longer than the patient's brain, extracranial regions are also scanned by CT to generate an accurate mu-map. Therefore, for PET scanners with a long axial FOV, when the region of interest is smaller than the PET scanner FOV, the current method results in the need to CT scan areas that are not clinically significant. This means that using a long FOV PET scanner exposes the patient to a higher dose of CT radiation, as well as organs adjacent to the region of interest, even if those organs are not the subject of clinical testing.

[0005] Therefore, there is a need in the art for improved methods for combined PET and CT scans. It would be particularly beneficial to provide a method for combined PET and CT scans that eliminates the need for CT scans of areas outside the region of interest. Summary of the Invention

[0006] According to one aspect of this disclosure, a method for minimizing patient exposure to CT scan radiation during the mu-map generation process in a PET scan is disclosed. The disclosed method is useful in PET scanners whose axial field of view (FOV) is particularly long compared to the volume of interest (VOI) in the patient. The method includes: performing a full-axial FOV PET scan of the patient and generating PET data; performing a truncated FOV CT scan of the VOI in the patient's body where the organ of interest is located; generating a truncated mu-map covering the truncated FOV of the CT scan, wherein the truncated FOV of the CT scan is shorter than the full-axial FOV of the PET scan; generating truncated PET data corresponding to the truncated mu-map, and using the truncated PET data to reconstruct a PET image of the VOI; and generating a mu-map for the full-axial FOV of the PET scan by extending the truncated mu-map generated from the truncated FOV CT scan by means of estimating missing mu-map data using the PET data. Attached Figure Description

[0007] The following will be apparent from the elements in the accompanying drawings, which are provided for illustrative purposes and are not necessarily drawn to scale.

[0008] Figure 1AThe illustration shows a nuclear imaging system for sequentially performing PET and CT scans, which can be used according to the method of this disclosure.

[0009] Figure 1B The illustration shows a computer system configured to implement one or more embodiments of the methods disclosed herein.

[0010] Figure 2 This is a flowchart outlining the method according to this disclosure.

[0011] Figure 3A The image shows a PET scan projection of a clinical cardiac scan.

[0012] Figure 3B The list pattern data shown is from areas outside the VOI-based CT scan after it has been turned off. Figure 3A The PET scan projection of the same patient.

[0013] Figure 4A It is a mu-image of a CT scan with the same axial scan length as the PET scan used, wherein the axial length is 159 slices of the CT image.

[0014] Figure 4B This refers to a mu-map for a truncated CT scan, which is generated by simulating a truncated CT scan with a shorter axial length compared to a single-bed PET scan by removing 10 slices from the top and 30 slices from the bottom of the CT image.

[0015] Figure 5A The crystal efficiency of the PET scanner is shown (798x80 in this example).

[0016] Figure 5B This shows what happens when the crystal outside the CT FOV is turned off. Figure 5A The crystal efficiency of the PET scanner.

[0017] Figure 6A Transaxial, coronal, and sagittal sections are shown from an ideal CT scan corresponding to a Vision (Siemens Healthineers) bed with an axial length of 159 slices.

[0018] Figure 6B Transverse, coronal, and sagittal slices of the TOF back-projected image obtained by norm correction are shown.

[0019] Figure 6C It shows Figure 5BThe image is segmented to generate a mask for detecting the support portion of the mu-graph in the truncated region.

[0020] Figure 7 The following are shown as transverse, coronal, and sagittal slices of reconstructed images obtained by using: (a) a reconstructed image using an untruncated sine plot and an untruncated mu-plot; (b) a reconstructed image using a truncated mu-plot and a truncated sine plot that has been rebinded to the same axial length as the truncated mu-plot; (c) a reconstructed image using a truncated mu-plot and an untruncated original sine plot; and (d) a reconstructed image using an extended mu-plot and an untruncated original sine plot.

[0021] Figure 8 The x-axis, coronal, and sagittal sections of the reconstructed images are shown. The scales for all images range from 0 to 50%. In (a), |bias|% is present in the reconstructed images using both truncated sine and truncated mu-plots. The voxel-by-voxel variation is primarily due to higher statistical variation in the reconstructed images. In (b), lower |bias|% is observed in the reconstructed images using both truncated mu-plots and the untruncated original sine plot (along all 80 rings along the z-axis). In (c), the minimum |bias|% is present in the reconstructed images using both the calculated extended mu-plot and the untruncated original sine plot. Within the myocardial region, |bias|% is less than 2%.

[0022] Figure 9A This is a flowchart outlining the method according to the present disclosure, wherein the proposed method is used when multiple organs are sampled by CT scans that are independent of each other. The PET scan can be a long-axis FOV scan, multiple step scans, or multiple continuous bed motion (CBM) scans.

[0023] Figure 9B This is a flowchart outlining the method according to the invention, wherein the proposed method is used to improve the scattering derived when the estimated extended mu-map is used.

[0024] Figure 9C This is a flowchart outlining the method according to the present disclosure, wherein the proposed method is used to improve scanner parameters, such as timing resolution, etc.

[0025] Figure 10A The illustration shows a complete mu-image of the patient generated from a CT scan, taken from head to thigh.

[0026] Figure 10B It shows Figure 10AAnother mu-image of the patient, which was generated from three separate truncated CT scans: the first was a truncated CT scan of the patient's brain only, the second was a truncated CT scan of the patient's heart region only, and the third was a truncated CT scan of the patient's pelvic region only.

[0027] Figure 11 This is a flowchart outlining embodiments of the methods of this disclosure, wherein the proposed methods are used to generate multiple sets of list patterns or sinusoids, which may or may not be limited to the area covered by CT. The maximum ring difference (MRD) and span angle used may be different for different list patterns or sinusoids and for the entire PET FOV sinusoid.

[0028] Figure 12 This is a flowchart outlining embodiments of the methods of this disclosure, wherein the methods of this disclosure are used to reconstruct PET images into a closed-form equation or into a sum of multiple reconstructions. It illustrates that if a user wishes to perform a CT scan on another area of ​​a patient's body while the patient is still in bed, new information from the CT scan can be used in conjunction with previously acquired data. Detailed Implementation

[0029] This description of exemplary embodiments is intended to be read in conjunction with the accompanying drawings, which are considered an integral part of the entire written description.

[0030] This paper discloses a method that allows generating mu-maps for corresponding PET scans by using a CT scan whose field of view (FOV) is truncated to be limited to the organ of interest, and then expanding the truncated mu-map by estimating mu-maps of regions outside the truncated CT FOV but still within the PET scan range using TOF PET data. In other words, the CT scan used to generate the mu-map is truncated to the region where the organ of interest resides, minimizing the area of ​​the patient's body exposed to radiation during the CT scan. However, this results in a truncated mu-map that lacks data on regions outside the CT FOV but still within the PET scan range. Therefore, the method disclosed herein expands the truncated mu-map by estimating the missing mu-map data using TOF PET data.

[0031] Figure 1AAn embodiment of a nuclear imaging system 2 in which the methods of this disclosure can be implemented is illustrated. The nuclear imaging system 2 includes at least a first imaging mode 12 provided in a first gantry 16a. The first imaging mode 12 may include any suitable mode, such as, for example, computed tomography (CT) mode, positron emission tomography (PET) mode, single-photon emission computed tomography (SPECT) mode, etc. The first imaging mode 12 may include a long-axis field of view (FOV) or a short-axis field of view (FOV). A patient 17 lies on a movable patient bed 18 movable relative to the first gantry 16a. In some embodiments, the nuclear imaging system 2 includes a second imaging mode 14 provided in a second gantry 16b. The second imaging mode 14 may be any suitable imaging mode, such as, for example, CT mode, PET mode, SPECT mode, and / or any other suitable imaging mode. The second mode 14 may include a long-axis FOV or a short-axis FOV. Each of the first imaging mode 12 and / or the second imaging mode 14 may include one or more detectors 50 arranged, for example, in one or more rings. Each of the detectors 50 is configured to detect annihilation photons, gamma rays, and / or other nuclear imaging events.

[0032] Scan data from the first imaging mode 12 and / or the second imaging mode 14 are stored in one or more computer databases 40 and processed by one or more computer processors 60 of the computer system 30. The graphical depiction of the computer system 30 in FIG1 is provided by way of illustration only, and the computer system 30 may include one or more separate computing devices, for example, as per [reference to...]. Figure 2 As described. Scan data may be provided by the first imaging mode 12, the second imaging mode 14, and / or may be provided as separate datasets, such as from memory coupled to the computer system 30. The computer system 30 may include one or more processing electronics for processing signals received from one of the plurality of detectors 50.

[0033] Figure 1B A computer system 30, configured to implement one or more processes according to some embodiments, is illustrated. System 30 is a representative device and may include a processor subsystem 72, an input / output subsystem 74, a memory subsystem 76, a communication interface 78, and a system bus 80. In some embodiments, one or more of the components of system 30 may be combined or omitted, such as, for example, omitting the input / output subsystem 74. In some embodiments, system 30 may include... Figure 1B Other components not shown. For example, system 30 may also include, for example, a power subsystem. In other embodiments, system 30 may include... Figure 1BSeveral instances of the components shown. For example, system 30 may include multiple memory subsystems 76. For the sake of brevity and clarity, and not limitation, Figure 1B One of each component is shown in the image.

[0034] The processor subsystem 72 may include any processing circuitry operable to control the operation and performance of the system 30. In various respects, the processor subsystem 72 may be implemented as a general-purpose processor, a multi-processor on a chip (CMP), a special-purpose processor, an embedded processor, a digital signal processor (DSP), a network processor, an input / output (I / O) processor, a media access control (MAC) processor, a radio baseband processor, a coprocessor, a microprocessor such as a Complex Instruction Set Computer (CISC) microprocessor, a Reduced Instruction Set Computing (RISC) microprocessor, and / or a Very Long Instruction Word (VLIW) microprocessor, or other processing device. The processor subsystem 72 may also be implemented by a controller, a microcontroller, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a programmable logic device (PLD), etc.

[0035] In various ways, the processor subsystem 72 can be configured to run an operating system (OS) and various applications. Examples of OSs include those commonly known by commercial names such as Apple OS, Microsoft Windows OS, Android OS, Linux OS, and any other proprietary or open-source OS. Examples of applications include, for example, network applications, local applications, data input / output applications, user interaction applications, etc.

[0036] In some embodiments, system 30 may include a system bus 80 coupling various system components, including a processing subsystem 72, an input / output subsystem 74, and a memory subsystem 76. System bus 80 may be any of several types of bus architectures, including memory buses or memory controllers, peripheral buses or external buses, and / or local buses using any of the various available bus architectures, including but not limited to 9-bit buses, Industry Standard Architecture (ISA), Micro Channel Architecture (MSA), Extended ISA (EISA), Intelligent Drive Electronics (IDE), VESA Local Bus (VLB), Peripheral Component Interconnect Card International Association Bus (PCMCIA), Small Computer Interface (SCSI) or other proprietary buses, or any custom bus suitable for computing device applications.

[0037] In some embodiments, the input / output subsystem 74 may include any suitable mechanism or component that enables a user to provide input to the system 30 and enables the system 30 to provide output to the user. For example, the input / output subsystem 74 may include any suitable input mechanism, including but not limited to buttons, keypads, keyboards, clickwheels, touchscreens, motion sensors, microphones, cameras, etc.

[0038] In some embodiments, the input / output subsystem 74 may include a visual peripheral output device for providing a display visible to a user. For example, the visual peripheral output device may include a screen, such as a liquid crystal display (LCD) screen. As another example, the visual peripheral output device may include a movable display or projection system for providing content display on a surface remote from system 30. In some embodiments, the visual peripheral output device may include an encoder / decoder, also referred to as a codec, for converting digital media data into analog signals. For example, the visual peripheral output device may include a video codec, an audio codec, or any other suitable type of codec.

[0039] Visual peripheral output devices may include display drivers, circuitry for driving the display drivers, or both. Visual peripheral output devices may be operable to display content under the direction of processor subsystem 72. For example, a visual peripheral output device may be able to play media playback information, application screens for applications implemented on system 30, information about ongoing communication operations, information about incoming communication requests, or device operation screens, to name just a few.

[0040] In some embodiments, the communication interface 78 may include any suitable hardware, software, or a combination of hardware and software capable of coupling the system 30 to one or more networks and / or additional devices. The communication interface 78 may be arranged to operate using any suitable technology for controlling information signals using a desired set of communication protocols, services, or operational procedures. The communication interface 78 may include appropriate physical connectors for connection to a corresponding communication medium, whether wired or wireless.

[0041] The carriers of communication include networks. In various respects, networks can include local area networks (LANs) and wide area networks (WANs), including but not limited to the Internet, wired channels, wireless channels, communication devices including telephones, computers, wired, radio, optical, or other electromagnetic channels, and combinations thereof, including other devices and / or components capable of transmitting data and associated with data transmission. For example, communication environments include intra-body communication, various devices, and various modes of communication, such as wireless communication, wired communication, and combinations thereof.

[0042] Wireless communication modes include any communication mode between points (e.g., nodes) that at least partially utilize wireless technology, which includes various protocols and combinations of protocols associated with wireless transmissions, data, and devices. These points include, for example, wireless devices such as wireless headsets, audio and multimedia devices and equipment such as audio players and multimedia players, telephones including mobile phones and cordless phones, and computers and computer-related devices and components such as printers, network-connected machinery, and / or any other suitable devices or third-party devices.

[0043] Wired communication modes encompass any communication mode between points utilizing wired technology, which includes various protocols and combinations of protocols associated with wired transmission, data, and devices. These points include devices such as: audio and multimedia devices and equipment such as audio players and multimedia players; telephones, including mobile phones and cordless phones; and computers and computer-related devices and components such as printers, network-connected machinery, and / or any other suitable devices or third-party devices. In various implementations, wired communication modules can communicate according to many wired protocols. Examples of wired protocols include Universal Serial Bus (USB) communication, RS-232, RS-422, RS-423, RS-485 serial protocols, FireWire, Ethernet, Fibre Channel, MIDI, ATA, Serial ATA, PCI Express, T-1 (and its variants), Industry Standard Architecture (ISA) parallel communication, Small Computer System Interface (SCSI) communication, or Peripheral Component Interconnect (PCI) communication, to name just a few.

[0044] Therefore, in various aspects, communication interface 78 may include one or more interfaces, such as, for example, wireless communication interfaces, wired communication interfaces, network interfaces, transmitting interfaces, receiving interfaces, media interfaces, system interfaces, component interfaces, switching interfaces, chip interfaces, controllers, etc. For example, when implemented by a wireless device or within a wireless system, communication interface 78 may include a wireless interface that includes one or more antennas, transmitters, receivers, transceivers, amplifiers, filters, control logic, etc.

[0045] In various aspects, communication interface 78 can provide data communication functionality according to a number of protocols. Examples of protocols may include various wireless local area network (WLAN) protocols, including the Institute of Electrical and Electronics Engineers (IEEE) 802.xx series of protocols, such as IEEE 802.11a / b / g / n / ac, IEEE 802.16, IEEE 802.20, etc. Other examples of wireless protocols may include various wireless wide area network (WWAN) protocols, such as GSM cellular radiotelephone system protocols with GPRS, CDMA cellular radiotelephone communication systems with 1xRTT, EDGE systems, EV-DO systems, EV-DV systems, HSDPA systems, etc. Further examples of wireless protocols may include wireless personal area network (PAN) protocols, such as infrared protocols, protocols from the Bluetooth Special Interest Group (SIG) series of protocols (e.g., Bluetooth specification versions 5.0, 6, 7, legacy Bluetooth protocols, etc.), and one or more Bluetooth profiles, etc. Yet another example of wireless protocols may include near-field communication technologies and protocols, such as electromagnetic induction (EMI) technology. Examples of EMI technologies can include passive or active radio frequency identification (RFID) protocols and devices. Other suitable protocols may include ultra-wideband (UWB), digital office (DO), digital home, trusted platform module (TPM), ZigBee, etc.

[0046] In some embodiments, at least one non-transitory computer-readable storage medium having computer-executable instructions embodied thereon is provided, wherein when executed by at least one processor, the computer-executable instructions cause at least one processor to perform embodiments of the methods described herein. The computer-readable storage medium may be embodied in memory subsystem 76.

[0047] In some embodiments, memory subsystem 76 may include any non-transitory machine-readable or computer-readable medium capable of storing data, including both volatile / non-volatile memory and removable / non-removable memory. Memory subsystem 8 may include at least one non-volatile memory cell. The non-volatile memory cell is capable of storing one or more software programs. The software program may contain, for example, applications, user data, device data and / or configuration data, or combinations thereof, to name just a few. The software program may contain instructions executable by various components of system 30.

[0048] In various respects, the memory subsystem 76 may include any non-transitory machine-readable or computer-readable medium capable of storing data, including both volatile / non-volatile memory and removable / non-removable memory. For example, the memory may include read-only memory (ROM), random access memory (RAM), dynamic RAM (DRAM), dual data rate DRAM (DDR-RAM), synchronous DRAM (SDRAM), static RAM (SRAM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory (e.g., NOR or NAND flash memory), content-addressable memory (CAM), polymer memory (e.g., ferroelectric polymer memory), phase-change memory (e.g., ovonic memory), ferroelectric memory, silicon-oxide-nitride-oxide-silicon (SONOS) memory, disk memory (e.g., floppy disk, hard disk, optical disk, magnetic disk), or card (e.g., magnetic card, optical card), or any other type of medium suitable for storing information.

[0049] In one embodiment, the memory subsystem 76 may contain a set of instructions in file form for performing various methods, such as those described herein, including A / B testing and cache optimization. The instruction set may be stored in any acceptable form of machine-readable instructions, including source code or various suitable programming languages. Some examples of programming languages ​​that may be used to store the instruction set include, but are not limited to, Java, C, C++, C#, Python, Objective-C, Visual Basic, or .NET programming. In some embodiments, a compiler or interpreter is included to translate the instruction set into machine-executable code for execution by the processing subsystem 72.

[0050] Figure 2Flowchart 100 outlines an embodiment of the method. First, a full-axial FOV PET scan is performed on the patient, thereby generating PET sinogram data (see step 110). A full-axial FOV scan can refer to one bed location or multiple bed locations. The method then includes performing a truncated FOV CT scan of the region in the patient's body containing the organ of interest (also referred to as the volume of interest (VOI) in the patient) (see step 120). Next, the method includes generating a truncated mu-map covering the truncated FOV of the CT scan, wherein the truncated FOV of the CT scan is shorter than the full-axial FOV of the PET scan (see step 130). In this embodiment, the truncated CT FOV is entirely within the single FOV of the PET scan. Because the mu-map generated from the truncated FOV CT scan does not match the full FOV of the long-axial FOV PET scan TOF data, data is missing from the mu-map data from the truncated FOV CT scan before the mu-map for the full PET FOV can be generated. Therefore, the method of this disclosure includes: expanding a truncated mu-map generated from a truncated FOV CT scan by estimating missing mu-map data using the entire measured TOF PET data from step 110, thereby generating a mu-map for the full-axial FOV of the PET scan (see step 140). The method may further include: reconstructing a PET image using the estimated mu-map for the full-axial FOV of the PET scan (i.e., the mu-map generated in step 140) (see step 150). Reconstruction step 150 includes: assigning different weights to the information content of different regions from the estimated mu-map during the PET image reconstruction process. Different weights are assigned to the portion of the full-axial FOV mu-map generated from the measured truncated FOV CT scan and the portion of the full-axial FOV mu-map generated by estimating the missing mu-map data of the region between the VOI scanned by the truncated FOV CT scan (i.e., the region not scanned by CT).

[0051] This method would be ideal for situations such as cardiac imaging, where the cardiac region is located at the center of a single-bed FOV, and CT is used to scan only the cardiac region, while the mu-values ​​of the rest of the body are jointly estimated using PET data. Some examples of other VOI imaging where this concept of organ-specific CT scanning can be used are breast scans, brain / prostate / pancreas / liver imaging.

[0052] Estimating the missing mu-map data in step 140 may include a combination of algorithms using prior prediction, numerical methods, CT reconnaissance scans, or artificial intelligence types. In the examples discussed herein, CT is a modality for obtaining anatomical information, and embodiments in which anatomical information is obtained by using MRI, ultrasound, or any other imaging modality or any combination of different modalities are included within the scope of this disclosure.

[0053] Estimating the missing mu-map data in step 140 may further include: calculating the average TOF PET emission value within the truncated FOV of the CT scan and the average TOF PET emission value outside the truncated FOV of the CT scan, and using the average TOF PET emission value to segment the norm-corrected PET image by identifying voxels above uptake thresholds for fat, muscle, and lung, and generating a mask for detecting the supporting portion of the mu-map in the region of the truncated FOV of the CT scan. This is illustrated below with some examples.

[0054] The process of generating a mu-map for full-axial FOV from a truncated mu-map from a truncated FOV CT scan will now be described in more detail. Figure 3A The image shows a PET scan projection sinusoid of a clinical cardiac scan obtained using a Siemens Biograph Vision scanner. The projection data consists of 520 pixels along the radial direction, 50 angle views, 33 TOF time bins, and 815 pixels along the axial direction. Figure 3B The list mode data showing the region outside the CT-based VOI has been turned off to simulate the truncated PET scan projection of the VOI corresponding to the truncated FOV CT scan of the VOI. Figure 3A The PET scan projection of the same patient. The list pattern data was reclassified to 520x815 (159, 139, 139, 101, 101, 63, 63, 25, 25) using a span of 19, yielding... Figure 3A and 3B The single-angle view projection data are shown. The patient was injected with 373 MBq of FDG and scanned 165 minutes post-injection. A two-minute step-and-scan (S&S) scan was performed on the cardiac region to study uptake in the myocardial region. The list pattern data were reclassified using a TOF smashing factor of 8 against a maximum ring difference (MRD) of 79.

[0055] Figure 4A The patient's mu-plot is shown in transverse, coronal, and sagittal views. Figure 4A The axial length of the mu-graph in the figure is related to the length used to generate the figure. Figure 3AThe axial length (26.1 cm) of the Biograph Vision scanner was the same for the PET scan projection. Then, to investigate the effect of using a shorter truncated FOV CT scan, 30 slices were removed from the bottom of the CT image and 10 slices were removed from the top of the CT image. This was done by closing the list pattern data of the region outside the VOI-based CT scan. The resulting PET scan projection sine curve... Figure 3B As shown in the diagram. Therefore, Figure 3B The sine wave shown illustrates what PET projection data would look like when matched to truncated FOV CT. This simulation was performed for a truncated FOV CT scan with a shorter axial length compared to a single-bed PET scan.

[0056] Figure 4A This is a mu-image of a CT scan corresponding to the axial scan length of a PET scanner with an axial length of 159 slices. Figure 4B This is a mu-image of a truncated FOV CT scan. The mu-image was generated by removing 10 data slices from the top and 30 data slices from the bottom of the CT image to simulate a truncated CT scan with a shorter axial length compared to a single-bed PET scan. The dark area 500 in the image is a result of the removed data.

[0057] Four methods were investigated for validation. The following were used to reconstruct PET data:

[0058] (1) The entire mu-plot (159 slices) (see Figure 4A ) and the entire PET sine curve used as the base reference (see Figure 3A );

[0059] (2) Mu-diagram of axial truncation ( Figure 4B ) and truncated, recombined PET sine maps axially matched to CT FOV ( Figure 3B );

[0060] (3) Mu-diagram of axial truncation ( Figure 4B ) and the entire PET sine curve (see Figure 3A );as well as

[0061] (4) Comparison with the estimated mu-plot (see Figure 6C The axially truncated mu-plot of the combination, and the entire sine plot ( Figure 3A ).

[0062] Figure 5AThe detector crystal efficiency of a PET scanner is shown. This particular example is from a clinical 8-ring BiographVision scanner (798x80). Crystal efficiency represents the detection efficiency of each crystal in the scanner for incident photons. For illustrative purposes, Figure 5A The illustration shows the crystal efficiency of the detector rings of a PET scanner, where the rings have been unfolded to be flat. To generate a truncated sine curve matching truncated FOV CT, the response line (LOR) outside the CT scan range was removed during the re-compartmentation step by modifying the scanner's crystal efficiency. This will be referenced below. Figure 5B This can be explained by generating a truncated sine wave that perfectly matches the truncated CT scan. Theoretically, a converged image can be reconstructed without any bias. The effect of missing data is modeled in norm during reconstruction.

[0063] exist Figure 5B In the diagram, the detector crystal efficiency is turned off for detector crystals outside the CT axial FOV. Missing data from the detector crystal ring outside the CT axial FOV is shown along... Figure 5B The top and bottom dark bands are 600. By disabling crystal efficiency, the LOR (Location of Circulation) of the area (area not measured by CT) is not considered during reconstruction. In the illustrated example, the bands of the 15 detector crystal rings at the top and the 5 detector crystal rings at the bottom of the scanner are disabled to correspond to... Figure 4B The CT images shown contain 30 and 10 slices of truncated data.

[0064] Figure 6A The image shows transverse, coronal, and sagittal slices from a CT scan corresponding to a Biograph Vision scanner bed with an axial length of 159 slices. To determine the support portion (e.g., outline) of the mu-graph in the truncated region, the norm-corrected PET data was back-projected using a non-attenuation correction (NAC) algorithm. Figure 6B The image shows the resulting transverse, coronal, and sagittal slices of the corrected time-of-flight (TOF) back-projection image.

[0065] Missing sections of the mu-map were estimated by calculating the average TOF PET emission values ​​within the truncated field of view (FOV) of the CT scan and the average TOF PET emission values ​​outside the truncated FOV of the CT scan. These average TOF PET emission values ​​were used to segment sections from fat, muscle, and lung by identifying voxels below the uptake thresholds for fat, muscle, and lung. Figure 6B Norm-corrected PET images, and generated for detection Figure 6CThe mask for the support portion of the mu-map in the truncated region shown (e.g., the contour, or attenuation region can be a rough approximation of the region therein). Therefore, Figure 6C This is an extended mu-map, which now includes attenuation data for regions within the PET scan's field of view (FOV) but outside the CT's truncated FOV. Missing portions of the CT / attenuation map can be estimated using segmentation, numerical methods, PET data, or a combination of artificial intelligence (AI). Alternatively, another approach is to use reconnaissance scans obtained from CT images to estimate the contour and attenuation along the locus of interest (LOR). The reconnaissance scan can be a single view or multiple views. Information from the reconnaissance scan can be integrated with the estimation step to generate the estimated mu-map.

[0066] The initial support portion of the obtained mu-map is slightly larger than the true mu-map measured by CT. This initial support portion of the mu-map serves as an ideal starting point for advanced joint estimation algorithms (such as maximum likelihood decay active reconstruction (MLAA) and maximum likelihood decay correction factor (MLACF)) that can be combined with AI-based methods to determine the extended mu-map.

[0067] A modified ordinary Poisson maximum likelihood expectation maximization (OP-MLEM) algorithm for modeling TOF, along with a point spread function (PSF) (25 iterations and 1 subset), was used for PET reconstruction. The modified OP-MLEM update equation is given by the following equation (1):

[0068]

[0069] in λ n yes n The image after one iteration, Y It is measurement data with missing data that are accurately modeled during the TOF crushing step. BP TOF It is a TOF back projection, and FP TOF It is a forward projection of Time-of-Flight (TOF). R These are random events. S It's scattering. N It is a norm, and A It's attenuation, subscript " c "Represents complete data, and includes the subscript" v "" represents the virtual gantry generated corresponding to the organ-specific truncated CT FOV. This update equation uses multiple sine maps, multiple mu-maps, multiple scatterings, and multiple norms for the same image update. α(alpha) can be multidimensional. The equation above is an example of explaining the modified OP-MLEM update equation for reconstruction, where the truncated sine curve from the virtual gantry and the fully measured data are used together with the truncated mu-plot, and the extended mu-plot is used to generate the reconstructed image. It should be noted that the norm, random events, attenuation, and scattering sine curve of the virtual gantry based on the truncated CT data (Fig. 10) and the measured sine curve can be different and will be modeled accordingly for reconstruction. In some embodiments of this disclosure, the modified OP-MLEM update equation above can also be written as, but is not limited to, the following equation (2):

[0070]

[0071] when Figure 3B When the truncated sine plot shown is used, the sensitivity decrease due to the missing LOR is modeled during the norm expansion step in order to accurately compensate for the missing data. Figure 5B (The missing data is represented in the image).

[0072] Figure 7 Line (a) shows the use of complete data (i.e., Figure 3A The untruncation sine curve shown) and the untruncation mu-curve ( Figure 4A The reconstructed image (shown in the image) is used as a baseline to calculate the bias and variance in reconstructed images obtained using other methods. Figure 7 The use of line (b) is shown Figure 4B The truncated mu-map and the sections that have been recombined to the same axial length as the truncated mu-map. Figure 3B The reconstructed image is a truncated sinusoid. This image exhibits high statistical variation throughout the image, and noisy voxels are more pronounced at the edges of the organ-specific truncated axial CT FOV, as indicated by the arrows. The increased noise in the reconstructed image is due to the reclassification of the original data, resulting in the removal of the LOR (Location of Observation) passing outside the truncated short-axis CT FOV. This effect can also be seen in… Figure 8 This can be seen in the absolute percentage deviation image in row (a). Here, in Figure 7 line (b) and Figure 7 The absolute percentage deviation per voxel was calculated between the images shown in row (a). Furthermore, the lack of modeling for attenuation effects in regions outside the CT FOV resulted in high deviations at edge slices, such as... Figure 8 The examples in rows (a) and (b) are shown.

[0073] Reconstructing the image using a truncated mu-plot and an uncrunted original sine plot resulted in less noise and bias (which will reduce noise and bias). Figure 7 line (c) and Figure 7(Compare with row (b)). Note that because the entire sine wave is used in the reconstruction, less noise and fewer artifacts are seen at the edges of the CT FOV. The robustness of the reconstructed image stems from the fact that, at the improved TOF timing resolution of 214 picoseconds, the effects of mismatches in the mu-image are very localized, and therefore the error does not propagate excessively into the PET reconstructed image (within the CT-sampled area). Finally, Figure 7 Row (d) shows the use of the extended mu-graph ( Figure 6C (as shown) and the untrunculated original sine curve ( Figure 3A The reconstructed image obtained. (e.g.) Figure 7 The percentage deviation shown in row (c) was found to be minimal when the extended mu-plot was used, and the deviation in the myocardial region was found to be less than 2%. Figure 7 The arrow in line (b) indicates the high-intensity voxels due to the truncation of the sine plot and the lower counts measured from that region, while Figure 7 The arrows in rows (c) and (d) show that the values ​​in those areas are more accurate and have less noise because more PET data was used during reconstruction.

[0074] By using organ-specific truncated FOV CT scans (one or more) with an axial length shorter than the published full-axial FOV PET scan, the radiation dose to organs outside the VOI during mu-image generation CT scans can be reduced. Furthermore, by recombining the raw PET data (full-axial FOV data), a virtual gantry with the same dimensions as the truncated short-axial FOV CT can be generated. Figure 3B and Figure 4B Because short-axis CT and PET data are perfectly matched, the reconstructed truncated sine wave data provides unbiased reconstructed images without any systematic artifacts.

[0075] According to some embodiments, the method of flowchart 100 can be... Figure 1A and 1B This is implemented in a nuclear imaging system 2. Such a system may include PET / CT scanner modalities 12 and 14, and a non-transitory machine-readable storage medium 76 that tangibly embodies an instruction program executable by a processor 60 to cause the processor to perform operations including:

[0076] (a) Perform a full-field-of-view (FOV) PET scan on the patient and generate PET data;

[0077] (b) Perform a truncated FOV CT scan of the volume of interest (VOI) in the patient's body;

[0078] (c) Generate a truncated mu-map covering the truncated FOV of the CT scan, where the truncated FOV of the CT scan is shorter than the full-axis FOV of the PET scan.

[0079] (d) Generate truncated PET data corresponding to the truncated mu-map, and use the truncated PET data to reconstruct a PET image of the VOI; and

[0080] (e) By using the PET data to estimate the missing mu-map data, the truncated mu-map generated from the truncated FOV CT scan is expanded, thereby generating a mu-map of the full FOV for the PET scan.

[0081] In some embodiments of this disclosure, the method of flowchart 100 can be applied to situations where all-axial FOV PET scanning is performed via multi-bed scanning or CBM scanning. This method is... Figure 9A The flowchart 200 shown is used to summarize this. In some embodiments, the method includes: performing a full-axial FOV PET scan of the patient (the full-axial FOV PET scan may be, but is not limited to, a single-bed scan, a multi-bed scan, or a CBM scan), which generates full-axial FOV PET scan data (see step 210); performing multiple truncated FOV CT scans of different regions in the patient's body where the organ of interest is located (see step 220); generating a truncated mu-map (Figure 10) covering the region scanned by the truncated CT FOV (see step 230); and generating a mu-map covering the region scanned by the truncated CT FOV by recombining the full-axial FOV PET scan data to match the finite axial length of the CT scan (i.e., the truncated FOV). The truncated PET sinusoidal data corresponding to the truncated mu-map of the scanned area of ​​FOV is used to reconstruct the PET image of the truncated area (see step 240); the PET image of the truncated area is reconstructed using the measured (i.e., truncated scan) CT data and the truncated PET sinusoidal data (see step 250); the mu-map is estimated using a combination of segmentation, numerical methods, PET data, or artificial intelligence (see step 260); and a whole-body (i.e., full PET FOV) PET image is reconstructed using a combination of truncated recombined PET scan data, truncated CT scan data, estimated CT data of the non-scanned area, other correction factors, and uncrunted measured PET scan data (see step 270).

[0082] According to some embodiments, the method of flowchart 200 can be... Figure 1A and 1BThis is implemented in a nuclear imaging system 2. Such a system may include PET / CT scanner modalities 12, 14 and a non-transitory machine-readable storage medium 76 that tangibly embodies an instruction program executable by a processor 60 to cause the processor to perform operations including the steps outlined in the flowchart 200 described above.

[0083] According to some embodiments, the method of flowchart 200 may further include the following steps: when the patient is in a patient bed, reconstructing a full FOV PET image using full-axial FOV PET scan data and a mu-map of the full-axial FOV for the PET scan; and if the full FOV PET image shows any aberrant uptake in any region not scanned by a truncated FOV CT scan, performing a truncated FOV CT scan of the region of aberrant uptake, and performing the following steps to reconstruct a PET image of the region of aberrant uptake: (i) generating a truncated mu-map from the truncated FOV CT scan of the region of aberrant uptake; (ii) generating truncated PET data corresponding to the truncated mu-map covering the region of aberrant uptake by recombining the PET scan data to match the truncated FOV of the truncated FOV CT scan; and (iii) using the truncated PET data to reconstruct a PET image of the region of aberrant uptake.

[0084] Furthermore, according to another embodiment of this disclosure, the truncated mu-plot generated in an embodiment of the process outlined in flowchart 100 or flowchart 200 can be used to improve scattering sine plot data obtained using the estimated mu-plot. Figure 9B This is a flowchart 300 summarizing this method for improving scattering sinogram data. It uses a truncated mu-map (i.e., a mu-map generated from a truncated FOVCT scan) (e.g., Figure 4B ) and truncated, recombined measured PET sine curve data (e.g., Figure 3B To generate the first scattering sine plot data (see step 310), the complete measured all-axial FOV PET sine plot data (e.g., with or without a truncated mu-plot) is then used. Figure 3A ) and the estimated mu-plot of the all-axial FOV (e.g., Figure 6CA second scattering sinusoidal plot is generated for the whole body (see step 320). Next, the second scattering sinusoidal plot data is compared with the first scattering sinusoidal plot data to determine if the second scattering sinusoidal plot data (whole-body scattering) is consistent with the first scattering sinusoidal plot data (see step 330). Since the mu-plot for the full-axis FOV includes the estimated portion, the scattering sinusoids from the estimated portion of that mu-plot may differ. If the second scattering sinusoidal plot data is inconsistent with the first scattering sinusoidal plot data, the first scattering sinusoidal plot data (from a truncated mu-plot (e.g., Figure 4B (scattering data) and truncated, recombined PET sine wave data (e.g.) Figure 3B The second scattering sinogram data (whole-body scattering data) is then improved or corrected (see step 340). This correction may include, but is not limited to, improving the scattering scale, improving the scattering shape, scattering modeling, and improving the overall quantization of the scattering sinogram obtained using the estimated mu-map. The improved / corrected scattering sinogram data can then be used to improve the estimation (i.e., re-estimation) of the mu-map of the patient's truncated FOV CT scan region (see step 350). The scattering from the re-estimated mu-map can then be used to improve the scattering from the truncated mu-map, such as, but not limited to, improving the scattering in edge slices of the truncated mu-map, and identifying scattering with smaller statistical changes.

[0085] refer to Figure 9C In flowchart 400, in some embodiments, the concept of truncated CT scanning can be used to detect and correct errors in scanner parameters. The method outlined in flowchart 400 can be used to check that the imaging parameters of a PET scanner are within tolerance. The method includes: firstly, performing a full-axial FOV PET scan on the patient to generate full-axial FOV PET data (see step 410). The full-axial FOV can be, but is not limited to, single-bed, multi-bed, or CBM scans. The method then includes: performing a truncated FOV CT scan for each of one or more VOIs (each VOI is a region in the patient's body containing an organ of interest) (see step 420). Next, generating a truncated mu-map from the truncated FOV CT scan data for each of the truncated FOV CT scan regions (i.e., VOIs) (see step 430). The measured CT regions (i.e., the truncated FOV CT scan regions) can be multiple regions of the body separated from each other; for example, a CT scan in one scan could be a CT scan of the brain, heart, and pelvis, with unsampled spatial regions in between. Examples of such truncated mu-maps are shown in... Figure 10B The diagram in the middle is shown. Figure 10A A complete mu-image of the patient, taken from head to thigh, is shown. Figure 10BA mu-map from a truncated CT scan that acquired only the brain, heart region, and pelvic region of the patient is shown. The method outlined in flowchart 400 further includes estimating a mu-map corresponding to the full-axial PET FOV, comprising one or more regions scanned by the truncated FOV CT scan, using full-axial FOV PET data (see step 440). Next, the method includes comparing the estimated / generated mu-map with the truncated mu-map (which is generated from the measured truncated FOV CT scan data) to see if the measured full-axial FOV PET data are consistent and free of any artifacts, such as time offset errors or gantry offsets (see step 450). If any mismatch is observed, a quality check of the PET scanner system can be performed. Furthermore, data obtained using the measured CT and PET data can be used to correct any inconsistencies in the parameters of the PET scanner system and to improve the measured PET scan data (see step 460). The improved PET scan data can now be used to re-estimate the mu-map in the truncated FOV CT scan region (i.e., VOI) of the patient (see step 470).

[0086] Another option is to use a truncated mu-plot generated from the measured truncated FOV CT scan and compare it with the estimated mu-plot to see if the PET scanner imaging parameters (e.g., time offset) are within the expected tolerance range. If the scanner imaging parameters are not within the expected tolerance range, a new calibration should be run for the PET scanner to update that calibration. Therefore, out-of-calibration status of the PET scanner can be determined without requiring a separate QC (quality control) study.

[0087] According to some embodiments, the method of flowchart 400 can be... Figure 1A and 1B This is implemented in a nuclear imaging system 2. Such a system may include PET / CT scanner modalities 12, 14, and a non-transitory machine-readable storage medium 76 that tangibly embodies an instruction program executable by a processor 60 to cause the processor to perform operations including the steps outlined in the flowchart 400 described above.

[0088] According to some embodiments, list pattern data from an all-axial FOV PET scan can be used directly in the methods of this disclosure, instead of PET sine curve data. Figure 11The flowchart 500 summarizes this concept. List pattern data 510, which identifies only the regions measured by the truncated FOV CT scan, is used to generate reconstructed PET images corresponding to the regions measured by the truncated FOV CT scan. In cases where there is more than one VOI and therefore multiple truncated FOV CT scans are performed (one truncated CT scan for each VOI), the same rules apply to each VOI region. In other words, for each VOI region, only the list pattern data passing through that region is identified, and it is used to generate a reconstructed PET image corresponding to that VOI region. The remaining list pattern data can be used to estimate missing mu-maps for regions not measured by the truncated FOV CT scan. After performing one or more organ-specific truncated FOV CT scans and generating corresponding truncated mu-maps (see 515) for each truncated FOV CT scan region, multiple sets of sine maps (see 520) are generated from the list pattern data 510 and the truncated mu-maps, one set for each CT scan region. Then, a sinogram matching the entire PET FOV (i.e., full-axis FOV PET) is generated (see 530). Based on the sinogram matching the entire PET FOV, the full FOV PET image can be reconstructed (see 540). Multiple list pattern data and sinograms may be limited to or not limited to the region measured by truncated FOV CT. For different sinograms and full PET FOV list pattern data and sinograms, the MRD and span angle used during reconstruction of the region measured by truncated FOV CT scan, as well as the estimated region, can be different from each other.

[0089] like Figure 12The flowchart 600 outlines the proposed method for reconstructing the final PET image using list pattern data and multi-organ-specific truncated CT scan data. This reconstruction uses sine wave / list pattern data matched to the full-axial FOV of the PET (see 610), multiple sine waves matched to the full-axial FOV (see 620), multi-organ-specific truncated FOV CT scans of the VOI (see 630), and correction factors (see 640). The reconstruction of the PET image can be performed as a closed-form equation or as the sum of multiple individual reconstructions (see 650). Furthermore, if clinicians observe any abnormalities in the PET image during the scan time, they can perform additional CT scans on only the abnormal areas (see step 660). For example, if a user wishes to perform a CT scan on another area of ​​the patient's body while the patient is still in bed, new information from the CT scan can be used in conjunction with previously acquired data and reconstructed. Scanner correction factors 640, such as norm, scattering, and random events, can be calculated at the same MRD and span angle as the PET data 610, and they can be different for various organ-specific CT scan regions.

[0090] According to another embodiment, a method is disclosed comprising: performing a full-axial FOV PET scan of a patient; performing a truncated FOV CT scan of the VOI; generating a mu-map of the CT scan area; estimating a mu-map of areas not scanned by the CT scan; while the patient is still in bed, reconstructing a full-axial FOV PET image using a mu-map that is a combination of the mu-map generated from the measured truncated FOV CT scan and the estimated mu-map; and if a clinician finds any abnormal uptake in any area not measured by the CT scan, a new truncated FOV CT scan can be performed only on the abnormal area, and a mu-map generated from the new truncated FOV CT scan; then reconstructing the PET image using an updated mu-map for the full-axial FOV, which now incorporates the mu-map generated from the new truncated FOV CT scan. This updated mu-map for the full-axial FOV is a combination of the mu-map generated from the measured new truncated FOV CT scan and the estimated mu-map.

[0091] The apparatus and processes are not limited to the specific embodiments described herein. Furthermore, components of each apparatus and each process may be practiced independently of and separately from the other components and processes described herein.

[0092] The preceding description of embodiments has been provided to enable any person skilled in the art to practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other embodiments without the use of inventive effort. This disclosure is not intended to be limited to the embodiments shown herein, but is accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method comprising the steps of: (a) performing a full axial field of view (FOV) PET scan of a patient and generating PET data; (b) performing a truncated FOV CT scan of a volume of interest (VOI) in the patient's body; (c) generating a truncated mu-map covering the truncated FOV of the CT scan, wherein the truncated FOV of the CT scan is shorter than the full axial FOV of the PET scan; (d) generating truncated PET data corresponding to the truncated mu-map and using the truncated PET data to reconstruct a PET image of the VOI; (e) generating a mu-map for the full axial FOV of the PET scan by extending the truncated mu-map generated from the truncated FOV CT scan by estimating missing mu-map data using the PET data; and (f) using the mu-map for the full axial FOV of the PET scan to reconstruct a PET image by assigning different weights to the information content from different regions in the mu-map for the full axial FOV of the PET scan.

2. The method of claim 1, wherein the PET scan is a single bed scan, a multi-bed scan, or a continuous bed motion scan.

3. The method of claim 1, wherein estimating the missing mu-map data comprises: using a combination of prior predictions, numerical methods, CT scout scans, or artificial intelligence type algorithms.

4. The method of claim 3, wherein estimating the missing mu-map data comprises: calculating average TOF PET emission values within the truncated FOV of the CT scan and average TOF PET emission values outside the truncated FOV of the CT scan and using the average TOF PET emission values to segment a norm corrected PET image by identifying voxels above uptake thresholds for fat, muscle, and lung and to generate a mask for detecting support of the mu-map in the truncated FOV of the CT scan.

5. A system comprising: a PET / CT scanner; a non-transitory machine readable storage medium tangibly embodying a program of instructions executable by a processor to cause the processor to perform operations comprising: (a) performing a full axial field of view (FOV) PET scan of a patient and generating PET data; (b) performing a truncated FOV CT scan of a volume of interest (VOI) in the patient's body; (c) generating a truncated mu-map covering the truncated FOV of the CT scan, wherein the truncated FOV of the CT scan is shorter than the full axial FOV of the PET scan; (d) generating truncated PET data corresponding to the truncated mu-map and using the truncated PET data to reconstruct a PET image of the VOI; (e) generating a mu-map for the full axial FOV of the PET scan by extending the truncated mu-map generated from the truncated FOV CT scan by estimating missing mu-map data using the PET data; and (f) using the mu-map for the full axial FOV of the PET scan to reconstruct a PET image by assigning different weights to the information content from different regions in the mu-map for the full axial FOV of the PET scan.

6. A non-transitory machine-readable storage medium tangibly embodying a program of instructions executable by a processor to cause the processor to perform operations comprising: (a) performing a full axial field of view (FOV) PET scan of a patient and generating PET data; (b) performing a truncated FOV CT scan of a volume of interest (VOI) in the patient's body; (c) generating a truncated mu-map covering the truncated FOV of the CT scan, wherein the truncated FOV of the CT scan is shorter than the full axial FOV of the PET scan; (d) generating truncated PET data corresponding to the truncated mu-map and using the truncated PET data to reconstruct a PET image of the VOI; (e) generating a mu-map for the full axial FOV of the PET scan by extending the truncated mu-map generated from the truncated FOV CT scan by estimating missing mu-map data using the PET data; and (f) reconstructing a PET image using the mu-map for the full axial FOV of the PET scan by assigning different weights to the information content from different regions in the mu-map for the full axial FOV of the PET scan.

7. A method comprising the steps of: (a) performing a full axial FOV PET scan of a patient and generating full axial FOV PET data; (b) performing a truncated field of view (FOV) CT scan for each of one or more volumes of interest (VOI) in which a organ of interest in the patient's body lies; (c) generating a truncated mu-map from each of the truncated FOV CT scans, wherein the truncated FOV of the CT scan is shorter than the full axial FOV of the PET scan; (d) generating truncated PET data corresponding to the truncated mu-map covering the VOI region scanned by the truncated FOV CT by re-binning the full axial FOV PET data to match the limited axial length of the CT scan and using the truncated PET data to reconstruct a PET image of the VOI region; (e) generating a mu-map for the full axial FOV of the PET scan by extending the truncated mu-map generated from each of the truncated FOV CT scans by estimating missing mu-map data for regions not covered by the truncated FOV CT scan using the full axial FOV PET data; and (f) reconstructing a full FOV PET image by assigning different weights to each of the truncated PET data generated for the truncated mu-maps and the full axial FOV PET data combined with the mu-map for the full axial FOV of the PET scan.

8. The method of claim 7, wherein the PET scan is a full axial FOV PET single bed scan, a multi-bed scan, or a continuous bed motion (CBM) scan, or a combination of a multi-bed scan and a CBM scan.

9. The method of claim 7, wherein estimating the mu-map data comprises: using a combination of prior predictions, numerical methods, CT scout scans, or artificial intelligence type algorithms, and using a combination of truncated PET data, truncated CT, estimated CT, other correction factors, and measured PET data to reconstruct the PET image.

10. The method of claim 7, further comprising: generating first scatter sinogram data using the truncated mu-map and the corresponding truncated PET data; generating second scatter sinogram data using the full axial FOV PET data and a mu-map for a full axial FOV of the PET scan; comparing the second scatter sinogram data to the first scatter sinogram data to determine whether the second scatter sinogram data is consistent with the first scatter sinogram data; in the event that the second scatter sinogram data is not consistent with the first scatter sinogram data, using the first scatter sinogram data and the truncated binned PET sinogram data to correct the second scatter sinogram data; and using the corrected second scatter sinogram data to re-estimate a mu-map for a truncated FOV CT scan region of the patient.

11. The method of claim 10, further comprising: using the re-estimated mu-map for the truncated FOV CT scan region of the patient to improve scatter correction for the truncated mu-map.

12. The method of claim 7, further comprising: reconstructing a full FOV PET image using the full axial FOV PET scan data and a mu-map for a full axial FOV of the PET scan while the patient is on the patient bed; and if the full FOV PET image exhibits any abnormal uptake in any region that was not scanned by the truncated FOV CT scan, performing a truncated FOV CT scan of the region of abnormal uptake and performing the following: generating a truncated mu-map from the truncated FOV CT scan of the region of abnormal uptake; generating truncated PET data corresponding to the truncated mu-map that covers the region of abnormal uptake by binning the PET scan data to match a truncated FOV of the truncated FOV CT scan; and reconstructing a PET image of the region of abnormal uptake using the truncated PET data.

13. A system comprising: a PET / CT scanner; a non-transitory machine-readable storage medium tangibly embodying a program of instructions executable by a processor to cause the processor to perform operations comprising: (a) performing a full axial FOV PET scan of a patient and generating full axial FOV PET data; (b) performing a truncated field of view (FOV) CT scan for each of one or more volumes of interest (VOIs) in which an organ of interest in the patient's body is located; (c) generating a truncated mu-map from each of the truncated FOV CT scans, wherein the truncated FOV of the CT scans is shorter than the full axial FOV of the PET scan; (d) generating truncated PET data corresponding to the truncated mu-maps covering the VOI region scanned by the truncated FOV CT by re-binning the PET data to match the limited axial length of the CT scan, and using the truncated PET data to reconstruct a PET image of the VOI region; and (e) generating mu-maps for the full axial FOV of the PET scan by extending the truncated mu-maps generated from each of the truncated FOV CT scans by estimating missing mu-map data for regions not covered by the truncated FOV CT scans using the full axial FOV PET data; and (f) reconstructing a full FOV PET image by assigning different weights to each of the truncated PET data generated for the truncated mu-maps and the full axial FOV PET data combined with the mu-maps for the full axial FOV of the PET scan.

14. A non-transitory machine-readable storage medium tangibly embodying a program of instructions executable by a processor to cause the processor to perform operations comprising: (a) performing a full axial FOV PET scan of a patient and generating full axial FOV PET data; (b) performing a truncated field of view (FOV) CT scan for each of one or more volumes of interest (VOIs) in which an organ of interest in the patient's body is located; (c) generating a truncated mu-map from each of the truncated FOV CT scans, wherein the truncated FOV of the CT scan is shorter than the full axial FOV of the PET scan; (d) generating truncated PET data corresponding to the truncated mu-maps covering the VOI region scanned by the truncated FOV CT by re-binning the PET data to match the limited axial length of the CT scan, and using the truncated PET data to reconstruct a PET image of the VOI region; and (e) generating mu-maps for the full axial FOV of the PET scan by extending the truncated mu-maps generated from each of the truncated FOV CT scans by estimating missing mu-map data for regions not covered by the truncated FOV CT scans using the full axial FOV PET data; and (f) reconstructing a full FOV PET image by assigning different weights to each of the truncated PET data generated for the truncated mu-maps and the full axial FOV PET data combined with the mu-maps for the full axial FOV of the PET scan.

Images