Photon-counting CT-guided pet image reconstruction

Abstract

Systems and methods include acquisition of PET data of an object, acquisition of multi-spectral CT data of the object, determination of a first subset of the multi-spectral CT data and a second subset of the multi-spectral CT data, reconstruction of a first CT image from the first subset of the multi-spectral CT data, reconstruction of a second CT image from the second subset of the multi-spectral CT data, and reconstruction of a PET image based on the PET data, the first CT image, and the second CT image.

Description

PHOTON-COUNTING CT-GUIDED PET IMAGE RECONSTRUCTIONBACKGROUND

[0001] Positron Emission Tomography (PET) generates quantitative images which represent biological processes (e.g., glucose metabolism, receptor affinity) occurring within a patient. Accordingly, PET images can help doctors diagnose and stage diseases, plan treatment, and evaluate the effectiveness of treatment. PET images are particularly useful in the evaluation of cancer, cardiovascular disease and brain disorders.

[0002] According to PET imaging, a radiotracer is administered to a patient via intravenous injection, inhalation, oral ingestion or direct organ injection. The tracer experiences radioactive decay as it travels within the patient, generating positrons which eventually encounter electrons and are annihilated thereby. An annihilation produces two 51 IkeV photons which travel in approximately opposite directions.

[0003] A ring of detectors surrounds the patient, and a coincidence is identified when two of the detectors detect the arrival of two photons within a short time window indicating that the two photons arose from the same positron annihilation. Because the two “coincident” photons travel in approximately opposite directions, the locations of the two detector crystals determine a Line-of-Response (LoR) along which an annihilation may have occurred. PET data represents each detected annihilation as a LoR between two detector crystals. Time-of- flight (TOF) PET additionally measures the difference between the detection times of the two photons arising from the annihilation. This difference may be used to estimate a particular position along the LOR at which the annihilation event occurred.

[0004] A three-dimensional PET image is reconstructed from the PET data using known algorithms such as filtered backproj ection (FBP) and ordered subsets expectation maximization (OSEM). PET image reconstruction is complex, resource-intensive and timeconsuming. Moreover, the resulting PET image typically exhibits low spatial resolution and significant noise, due in part to the size of the detectors and a low' number of detected coincidences.

[0005] Current approaches attempt to improve the quality of PET images by leveraging anatomical information within the PET reconstruction process. The anatomical informationmay consist of a map of linear attenuation coefficients determined from Computed Tomography (CT) data of the patient. However, CT data poorly delineates organ and other soft tissue boundaries and therefore the quality of a PET image reconstructed using such a linear attenuation coefficient map is often unsuitable. Magnetic Resonance (MR) images may be used to define hard and soft tissue boundaries which are used in subsequent PET reconstruction, but the availability of MR scanners is limited.

[0006] Systems are desired to efficiently improve spatial resolution and noise characteristics of PET images. Such systems preferably strike a desirable balance between image quality and radiation exposure.BRIEF DESCRIPTION OF THE DRAWINGS

[0007] FIG. 1 is a block diagram illustrating reconstruction of a PET image using two images generated from multi-spectral CT data according to some embodiments.

[0008] FIG. 2 is a block diagram illustrating operation of a photon-counting CT detector according to some embodiments.

[0009] FIG. 3 is a flow diagram of a process to reconstruct a PET image using two images generated from multi-spectral CT data according to some embodiments.

[0010] FIG. 4 is a block diagram illustrating reconstruction of a PET image using two images generated from multi-spectral CT data according to some embodiments.

[0011] FIG. 5 is a block diagram of an imaging system according to some embodiments.DETAILED DESCRIPTION

[0012] The following description is provided to enable any person in the art to make and use the described embodiments. Various modifications, however, will remain apparent to those in the art.

[0013] Embodiments utilize photon-counting CT to generate two or more CT images, hereinafter referred to as virtual monoenergetic images (VMI), from a single CT scan. The two or more images are used to guide reconstruction of a PET image from PET data. Each ofthe two or more images may represent different spectra of photon energies, and each of the two or more images may exhibit less noise and higher spatial resolution than a conventional CT image. Moreover, the use of two CT images of different spectra may improve soft tissue differentiation and result in more accurate attenuation correction values for use in PET reconstruction. Embodiments may thereby generate PET images with lower noise, higher spatial resolution and higher quantification accuracy in a practical timeframe and with less delivered radiation dose than current approaches.

[0014] FIG. 1 is a block diagram of system 100 to generate a corrected PET image according to some embodiments. The illustrated components of system 100 may be implemented in computer hardware, in program code and / or in one or more computing systems executing such program code as is known in the art. Such a computing system may include one or more processing units which execute program code stored in one or more non- transitory storage media. More than one functional component may be implemented by a single computing system in some embodiments. One or more of the computing systems may comprise a virtual machine, and one-or more computing systems may comprise a cloud-based compute resource providing on-demand scalability and failure recovery'.

[0015] Scanner 110 may comprise a PET / CT scanner capable of generating PET data and CT data associated with an object such as patient 115. As will be described below, embodiments are not limited to the use of a single scanner to produce PET data and CT data.

[0016] Scanner 110 uses a photon-counting detector to generate CT data 120. CT data 120 is acquired while an X-ray tube and the photon-counting detector of scanner 110 are positioned at various projection angles Po to Pnwith respect to patient 115. Generally, CT data 120 includes a set of data acquired at each of projection angles Po to Pn.

[0017] Each set of data indicates the number of photons detected at each pixel (xm. yn) of an M x N two-dimensional projection space while the tube and detector were positioned at the corresponding projection angle P. Moreover, the data indicates how many photons of each of four energy ranges (e.g., ~30 keV, ~45 keV. ~65 keV and ~90 keV) were detected at each pixel. Embodiments are not limited to four energy ranges.

[0018] As illustrated at FIG. 1, a set of data corresponding to a projection angle P may be characterized as a set of histograms associated with each detector pixel, where a histogram indicates how many photons having energies within each energy range were detected at thepixel at the projection angle P. In contrast to CT data 120, CT data acquired by a conventional CT detector integrates the energies of all photons received at a pixel to generate a single value per pixel at each projection angle.

[0019] Scanner 110 acquires PET data 130 using any suitable PET data acquisition protocol. As described above, a coincidence is identified when two PET detectors detect the arrival of two photons within a short time window. For each detected coincidence, PET data 130 provides the locations of the two PET detector crystals which detected the photons (i.e., and which define a LoR), the time at which each photon of the coincidence arrived at each PET detector crystal and, in some cases, the difference between the arrival times of the two photons (i.e.. ToF data).

[0020] The PET detector cry stals may comprise lutetium oxyorthosilicate (LSO), lutetiumyttrium oxy orthosilicate (LYSO), or any other suitable materials that are or become known. The crystals create light photons in response to receiving 511 keV photons and in response to receiving the emitted background radiation. Electrical transducers, or photosensors, convert these light photons to electrical signals, sometimes referred to herein as pulses. According to some embodiments, the electrical transducers may comprise silicon photomultipliers (SiPMs) or photomultiplier tubes (PMTs).

[0021] PET data 130 may represent the detected coincidences as raw (i.e., list-mode) data and / or sinograms. List-mode data may represent each coincidence using data specifying a LoR between two crystals, the time at which each photon of the annihilation reached each crystal, the photon energies, etc. A sinogram is a data array of the angle versus the displacement of the LoRs of each detected coincidence. A sinogram includes one row containing the LoR for a particular azimuthal angle <p. Each of these rows corresponds to a one-dimensional parallel projection of the tracer distribution at a different coordinate. A sinogram stores the location of the LoR of each coincidence such that all the LoRs passing through a single point in the volume trace a sinusoid curve in the sinogram.

[0022] Scanner 110 may generate CT data 120 and PET data 130 substantially contemporaneously. For example, a PET imaging system of scanner 110 may be operated to acquire PET data 130 while patient 115 lies in a given position on a bed of scanner 110, and a CT imaging system of scanner 110 may be operated shortly thereafter to acquire CT data 120 while the patient remains on the bed in the given position. Since the geometrictransformation (if any) between coordinates of the PET imaging system and the CT imaging system is known, CT data 120 and PET data 130 may be easily spatially registered with one another.

[0023] CT data 122 and CT data 124 are derived from CT data 120. According to some embodiments, CT data 122 and CT data 124 may represent different energy spectra. In the illustrated embodiment, CT data 122 represents detected photons within a single energy range and CT data 124 represents detected photons within a single energy range that is lower than the energy range of CT data 122.

[0024] The energy spectra represented by CT data 122 and CT data 124 may overlap such that some detected photons may be represented in both CT data 122 and CT data 124. The energy spectra may be discontinuous, where, for example, CT data 122 represents photons of both lower and higher energies than the photons of CT data 124. Embodiments are not limited to any particular criteria for generating CT data 122 and CT data 124 from CT data 120. Moreover, embodiments are not limited to the generation of two data sets from CT data 120.

[0025] Reconstruction components 140 and 145 respectively reconstruct CT images 150 and 155 from CT data 122 and CT data 124. Since CT data 122 and CT data 124 represent different energy spectra, CT images 150 and 155 may also be considered as representing different energy spectra. For example, CT image 150 may represent photons having energies in the range of 65 keV while CT image 155 may represent photons having energies in the range of 30 keV. CT image 150 represents attenuation of the photons of CT data 122 by various anatomical structures of patient 115, and CT image 155 represents attenuation of the photons of CT data 124 by the anatomical structures. CT image 150 and CT image 155 may therefore be considered anatomical images.

[0026] Reconstruction components 140 and 145 may use the same or different reconstruction algorithms. As is known in the art, a reconstruction algorithm may implement analytical reconstruction or iterative reconstruction, for example. The most commonly used analytical reconstruction algorithms utilize FBP. FBP applies a ID filter on the CT data and backprojects the filtered data onto the image space. A smoother ID filter (or, “kernel”) generates images with lower noise but with reduced spatial resolution, while a sharper kernel generates images with higher spatial resolution but increased noise. Iterative reconstructionalgorithms reconstruct images by iteratively optimizing an objective function, which typically consists of a data fidelity term and an edge-preserving regularization term. The optimization includes iterations of forward projection and backproj ection of the CT data between image space and projection space.

[0027] Reconstruction component 160 reconstructs PET image 170 from PET data 130.The reconstruction of PET image 170 utilizes CT image 150 and CT image 155.Embodiments are not limited to two CT images. For example, four sets of different CT data may be derived from CT data 120 and used to reconstruct four corresponding CT images, and the reconstruction of PET image 170 may utilize all four CT images.

[0028] According to some embodiments, reconstruction component 160 may employ an iterative statistical algorithm, such as OSEM, to generate PET image 170. The algorithm may use an objective function (i.e. , objective model) that includes a prior information function (i. e. , prior information model) with a similarity function (i.e.. similarity model) that applies weights based on values of multiple VMI: CT image 150 and CT image 155. In one example, the weights may be computed as follows:where j is the voxel index, b is the index for the voxels in the neighborhood around voxel j. Vi is CT image 150, v2is CT image 155, k is the number of VMI. o is the standard deviation in u and v, and p(u2v,) is defined as a non-parametric Parzen window using Gaussian kernels. This weighting can be extended to as many VMI as desired.

[0029] Due to the low noise, high spatial resolution and improved soft tissue differentiation of CT images 150 and 155, resulting PET image 170 may exhibit lower noise, higher spatial resolution, better differentiation between soft tissues, and higher quantification accuracy than previously available.

[0030] FIG. 2 illustrates acquisition of a CT projection image according to some embodiments. X-ray tube 210 is mounted on a cylindrical gantry (not shown) opposite photon-counting CT detector 230. The gantry is rotatable about object 220 such that X-raytube 210 and detector 230 may be positioned at various projection angles with respect to object 220.

[0031] At each projection angle, X-ray tube 210 emits X-rays (i.e., photons) toward object 220. The photons are attenuated by structures within object 220 and the attenuated photons then reach detector 230. The photons pass through cathode 232 and are absorbed by semiconductor 234 (e.g., cadmium telluride). The absorption creates electron-hole pairs and the electrons are directed to anode electrodes 236 by an electric field established between cathode 232 and electrodes 236.

[0032] Each electrode 236 represents a pixel and receives individual electrical charges of each photon received at the pixel. Unlike conventional scintillation-based detectors, detector 230 does not require septa between the pixels to avoid optical cross talk. Detector signal processing component 240 counts the charges created by individual photons at each pixel and measures their energy level to generate CT data for the projection angle as shown in FIG. 1. Moreover, component 240 may set a threshold energy below which electrode signals are ignored, ensuring that the only signals generated from valid photons are considered and that the resulting CT data exhibits a high signal-to-noise ratio.

[0033] FIG. 3 is a flow diagram of process 300 to reconstruct a PET image using two images generated from the multi-spectral CT data according to some embodiments. Embodiments are not limited to two CT images. Process 300 may be performed by any combination of hardware and software that is or becomes known. Program code embodying processes described herein may be stored by any one or more non-transitory tangible media, including a fixed disk, a volatile or non-volatile random-access memory, a DVD, a Flash drive, and a magnetic tape, and executed by any suitable processing unit, including but not limited to one or more microprocessors, microcontrollers, processor cores, and processor threads. Embodiments are not limited to the examples described below.

[0034] PET data of an object is acquired at S310. The PET data may be acquired using any suitable PET imaging system and any suitable PET imaging protocol. Typically, a radiotracer is injected to the object and acquisition of PET data begins after a certain time period has elapsed. The PET data may comprise list-mode data or sinogram data.

[0035] Multi-spectral CT data of the object is acquired at S320. The multi-spectral CT data may be acquired by performing a CT scan using a photon-counting detector. The multi-spectral CT data may include, for each of a plurality of projection angles and for each pixel of the photon-counting detector, a count of photons of each of two or more energy ranges.

[0036] First CT data is determined based on the multi-spectral CT data at S330. The first CT data may comprise a subset of the multi-spectral CT data associated w ith photons having energies within a first one or more energy ranges. Similarly, second CT data is determined based on the multi-spectral CT data at S340, for example consisting of a subset of the multi- spectral CT data which is associated with photons having energies within a second one or more energy ranges. The first one or more energy' ranges and the second one or more energy' ranges may partially overlap in some embodiments.

[0037] A first CT image is generated based on the first CT data at S350 using any suitable reconstruction algorithm. A second CT image is generated based on the second CT data at S360, also using any suitable reconstruction algorithm. The reconstruction algorithms used at S350 and S360 may be the same or different from one another.

[0038] At S370, a PET image is reconstructed based on the PET data acquired at S310, the first CT image and the second CT image. As described above, in some embodiments, the reconstruction at S370 employs an iterative statistical algorithm which uses an objective function that includes a similarity function which uses weights determined from the values of first CT image and the second CT image. The reconstructed PET image is displayed ay S380. The PET image may be displayed in three-dimensions, as slices, or in any other known manner.

[0039] FIG. 4 illustrates reconstruction of a PET image using two images generated from multi-spectral CT data according to some embodiments. FIG. 4 may represent an implementation of the system of FIG. 1 and / or process 300 according to some embodiments.

[0040] PET data 405 of an obj ect may' be acquired as described above with respect to PET data 130 of FIG. 1. PET data 405 may be acquired by a standalone PET scanner and / or a PET / CT scanner. PET data 410 is reconstructed using known techniques (e.g., non- attenuati on-corrected PET reconstruction techniques) to generate PET image 420.

[0041] CT image 430 and CT image 435 w ere generated from different subsets of multi- spectral CT data as described above. The multi-spectral CT data may have been generated by a single scan of a CT scanner including a photon-counting detector. The CT scanner may bea standalone CT scanner or a component of a PET / CT scanner which was also used to acquire PET data 405.

[0042] Registration component 440 registers CT image 430 and CT image 435 to PET image 420 to generate registered CT image 450 and registered CT image 455. Registration component 440 may utilize any suitable registration algorithm. As noted above, registration may be nominal in a case that PET data 405, CT image 430 and CT image 435 were generated using the same PET / CT scanner.

[0043] Reconstruction component 460 reconstructs PET image 470 based on PET data 405, registered CT image 450 and registered CT image 455. Reconstruction may include use of an objective function that considers values of registered CT image 450 and registered CT image 455.

[0044] FIG. 5 illustrates PET / CT scanner 500 to execute one or more of the processes described herein. Embodiments are not limited to scanner 500 or to a multi-modality imaging system.

[0045] Scanner 500 includes gantry 510 defining bore 512. As is known in the art. gantry 510 houses PET imaging components for acquiring PET image data and CT imaging components for acquiring CT image data. The CT imaging components may include one or more x-ray tubes and one or more corresponding photon-counting detectors. The PET imaging components may include any number or type of PET detectors disposed in any configuration as is known in the art.

[0046] Bed 515 and base 516 are operable to move a patient lying on bed 515 into and out of bore 512 before, during and after imaging. In some embodiments, bed 515 is configured to translate over base 516 and, in other embodiments, base 516 is movable along with or alternatively from bed 515.

[0047] Movement of a patient into and out of bore 512 may allow scanning of the patient using the CT imaging elements and the PET imaging elements of gantry 510. Bed 515 and base 516 may provide continuous bed motion and / or step-and-shoot motion during such scanning according to some embodiments.

[0048] Control system 520 may comprise any general-purpose or dedicated computingsystem. Accordingly, control system 520 includes one or more processing units 522 configured to execute program code to cause system 520 to acquire image data and generate images therefrom, and storage device 530 for storing the program code. Storage device 530 may comprise one or more fixed disks, solid-state random-access memory, and / or removable media (e.g., a thumb drive) mounted in a corresponding interface (e.g., a Universal Serial Bus port).

[0049] Storage device 530 stores program code of control program 531. One or more processing units 522 may execute control program 531 to control CT imaging elements of scanner 500 using CT system interface 524 and bed interface 525 to acquire multi-spectral CT data and to reconstruct CT images 534 from subsets thereof. One or more processing units 522 may execute control program 531 to, in conjunction with PET system interface 523 and bed interface 525, control hardware elements to inject a radiopharmaceutical into a patient, move the patient into bore 512 past PET detectors of gantry 510, and detect photons emitted from the patient based on pulses generated by the PET detectors. The detected photons may be recorded as PET data 533, which may be reconstructed into a PET image 535 using the CT images as described above.

[0050] PET images 535 and CT images 534 may be transmitted to terminal 540 via terminal interface 526. Terminal 540 may comprise a display device and an input device coupled to system 520. Terminal 540 may display the received PET images 535 and CT images 534. Terminal 540 may receive user input for controlling display of the data, operation of scanner 500, and / or the processing described herein. In some embodiments, terminal 540 is a separate computing device such as, but not limited to. a desktop computer, a laptop computer, a tablet computer, and a smartphone.

[0051] Each component of scanner 500 may include other elements which are necessary for the operation thereof, as well as additional elements for providing functions other than those described herein. Each functional component described herein may be implemented in computer hardware, in program code and / or in one or more computing systems executing such program code as is known in the art. Such a computing system may include one or more processing units which execute processor-executable program code stored in a memory system.

[0052] Those in the art will appreciate that various adaptations and modifications of the above-described embodiments can be configured without departing from the claims. Therefore, it is to be understood that the claims may be practiced other than as specifically described herein.

Claims

WHAT IS CLAIMED IS:1 . An imaging system comprising: an x-ray tube; a detector for receiving photons emitted by the x-ray tube and generating multi- spectral data; a plurality of positron emission tomography (PET) detectors for generating PET data based on photons generated by positron annihilations; a processing unit to: determine a first subset and a second subset of the multi-spectral data; reconstruct a first computed tomography (CT) image from the first subset of the multi-spectral data; reconstruct a second CT image from the second subset of the multi-spectral data; and reconstruct a PET image based on the PET data, the first CT image, and the second CT image; and a display to present the PET image.

2. The imaging system of Claim 1, wherein the detector is a photon-counting detector.

3. The imaging system of Claim 1 wherein the first subset represents photons associated with a first one or more energy7ranges and the second subset represents photons associated with a second one or more energy ranges.

4. The imaging system of Claim 3, wherein reconstruction of the PET image based on the PET data, the first CT image, and the second CT image comprises reconstruction of the PET image based on an objective function including a similarity function having weights based on voxel values of the first CT image and the second CT image.

5. The imaging system of Claim 1, wherein reconstruction of the PET image based on the PET data, the first CT image, and the second CT image comprises reconstruction of the PET image based on an objective function including a similarity’ function having weights based on voxel values of the first CT image and the second CT image.

6. The imaging system of Claim 5, wherein reconstruction of the PET image comprises: reconstruction of a first PET image from the PET data; registration of the first CT image and the second CT image to the first PET image; and reconstruction of the PET image based on the PET data, the registered first CT image, and the registered second CT image.

7. The imaging system of Claim 1, wherein reconstruction of the PET image comprises: reconstruction of a first PET image from the PET data; registration of the first CT image and the second CT image to the first PET image; and reconstruction of the PET image based on the PET data, the registered first CT image, and the registered second CT image.

8. A method comprising: acquiring positron emission tomography (PET) data of an object; acquiring multi-spectral computed tomography (CT) data of the object; determining a first subset of the multi-spectral CT data and a second subset of the multi-spectral CT data; reconstructing a first CT image from the first subset of the multi-spectral CT data; reconstructing a second CT image from the second subset of the multi-spectral CT data; and reconstructing a PET image based on the PET data, the first CT image, and the second CT image.

9. The method of Claim 8. wherein the multi-spectral CT data comprises a photoncounting detector.

10. The method of Claim 8, wherein the first subset represents photons associated with a first one or more energy ranges and the second subset represents photons associated with a second one or more energy ranges.

11. The method of Claim 10, wherein reconstructing the PET image based on the PET data, the first CT image, and the second CT image comprises reconstructing the PET image based on an objective function including a similarity function having weights based on voxel values of the first CT image and the second CT image.

12. The method of Claim 8, wherein reconstructing the PET image based on the PET data, the first CT image, and the second CT image comprises reconstructing the PET image based on an objective function including a similarity function having weights based on voxel values of the first CT image and the second CT image.

13. The method of Claim 12, wherein reconstructing the PET image comprises: reconstructing a first PET image from the PET data; registering the first CT image and the second CT image to the first PET image; and reconstructing the PET image based on the PET data, the registered first CT image, and the registered second CT image.

14. The method of Claim 8, wherein reconstructing the PET image comprises: reconstructing a first PET image from the PET data; registering the first CT image and the second CT image to the first PET image; and reconstructing the PET image based on the PET data, the registered first CT image, and the registered second CT image.

15. One or more non-transitory computer-readable media storing program code that, when executed by a computing system, causes the computing system to perform operations comprising: acquiring multi-spectral computed tomography (CT) data of an object; determining a first subset of the multi-spectral CT data and a second subset of the multi-spectral CT data; reconstructing a first CT image from the first subset of the multi-spectral CT data;reconstructing a second CT image from the second subset of the multi-spectral CT data; and reconstructing a PET image based on PET data, the first CT image, and the second CT image.

16. The one or more non-transitory computer-readable media of Claim 15, wherein the first subset represents photons associated with a first one or more energy ranges and the second subset represents photons associated with a second one or more energy ranges.

17. The one or more non-transitory computer-readable media of Claim 16, wherein reconstructing the PET image based on the PET data, the first CT image, and the second CT image comprises reconstructing the PET image based on an objective function including a similarity function having weights based on voxel values of the first CT image and the second CT image.

18. The one or more non-transitory computer-readable media of Claim 15, wherein reconstructing the PET image based on the PET data, the first CT image, and the second CT image comprises reconstructing the PET image based on an objective function including a similarity function having weights based on voxel values of the first CT image and the second CT image.

19. The one or more non-transitory computer-readable media of Claim 18. wherein reconstructing the PET image comprises: reconstructing a first PET image from the PET data; registering the first CT image and the second CT image to the first PET image; and reconstructing the PET image based on the PET data, the registered first CT image, and the registered second CT image.

20. The one or more non-transitory computer-readable media of Claim 15, wherein reconstructing the PET image comprises: reconstructing a first PET image from the PET data; registering the first CT image and the second CT image to the first PET image; and reconstructing the PET image based on the PET data, the registered first CT image, and the registered second CT image.

Images