Advances in SPECT and PET Hardware Piotr J. Slomka, Tinsu Pan, Daniel S. Berman, Guido Germano PII: DOI: Reference:

S0033-0620(15)00011-0 doi: 10.1016/j.pcad.2015.02.002 YPCAD 645

To appear in:

Progress in Cardiovascular Diseases

Please cite this article as: Slomka Piotr J., Pan Tinsu, Berman Daniel S., Germano Guido, Advances in SPECT and PET Hardware, Progress in Cardiovascular Diseases (2015), doi: 10.1016/j.pcad.2015.02.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Advances in SPECT and PET Hardware

T

Piotr J. Slomka, Ph.D.

RI P

Research Scientist, Artificial Intelligence Program, Cedars-Sinai Medical Center Professor of Medicine, UCLA School of Medicine

SC

Los Angeles, California 90048

NU

Ph: (310)423-4348, Email: [email protected], Fax: (310)423-0173

Tinsu Pan, PhD

MA

Professor of Imaging Physics, University of Texas, MD Anderson Cancer Center Houston, TX 77030

PT

Daniel S. Berman, M.D.

ED

Ph: (713)563-2714, Email: [email protected]

CE

Director, Cardiac Imaging, Cedars-Sinai Medical Center Professor of Medicine, UCLA School of Medicine

AC

Los Angeles, California 90048

Ph: (310)423-4223, Email: [email protected], Fax: (310)423-0173

Guido Germano, Ph.D. Director, Artificial Intelligence Program, Cedars-Sinai Medical Center Professor of Medicine, UCLA School of Medicine Los Angeles, California 90048 Ph: (310)423-4344, Email: [email protected], Fax: (310)423-0173

1

ACCEPTED MANUSCRIPT

T

Corresponding Author:

RI P

Piotr J. Slomka, Ph.D.

Research Scientist, Artificial Intelligence Program, Cedars-Sinai Medical Center

SC

Professor of Medicine, UCLA School of Medicine Los Angeles, California 90048

NU

Ph: (310)423-4348, Email: [email protected], Fax: (310)423-0173

ED

MA

Keywords: SPECT, PET, scintillation detectors, solid state detectors, digital detectors, cadmium zinc telluride, time-of-flight, PET/MR, PET/CT, collimators, cardiac imaging, radiation dose, cardiac PET, cardiac SPECT

PT

Disclosures

AC

ABSTRACT

CE

Piotr Slomka has received research grant support from Siemens and has received NIH grant support for a analysis of SPECT images from the new generation of SPECT scanners. Tinsu Pan has received research grant support from GE. Dan Berman has received grant support from Siemens.

There have been significant recent advances in single photon emission computed tomography (SPECT) and positron emission tomography (PET) hardware. Novel collimator designs, such as multi-pinhole and locally focusing collimators arranged in geometries that are optimized for cardiac imaging have been implemented to reduce imaging time and radiation dose. These new collimators have been coupled with solid state photon detectors to further improve image quality and reduce scanner size.

The new SPECT

scanners demonstrate up to a 7-fold increase in photon sensitivity and up to 2 times improvement in image resolution. Although PET scanners are used primarily for oncological imaging, cardiac imaging can benefit from the improved PET sensitivity of 3D systems without inter-plane septa and

2

ACCEPTED MANUSCRIPT implementation of the time-of-flight reconstruction. Additionally, resolution recovery techniques are now implemented by all major PET vendors. These new methods improve image contrast, image resolution,

T

and reduce image noise. Simultaneous PET/ magnetic resonance (MR) hybrid systems have been

RI P

developed. Solid state detectors with avalanche photodiodes or digital silicon photomultipliers have also been utilized in PET. These new detectors allow improved image resolution, higher count rate, as well as

SC

a reduced sensitivity to electromagnetic MR fields.

NU

Abbreviations:

APD-Avalanche photodiode detection

ED

CT-Computed tomography

MA

AC-Attenuation correction

PT

CV-Cardiovascular

CE

CZT-Cadmium-Zinc-Telluride

AC

LSO-Lutetium-oxyorthosilicate

LYSO-Lutetium-yttrium-orthosilicate LV-Left ventricular

MR-Magnetic resonance PET-Positron emission tomography SiPM-Silicon digital photomultipliers SPECT-Single photon emission computed tomography TOF-Time-of-flight 3

ACCEPTED MANUSCRIPT Introduction Single photon emission computed tomography (SPECT) and positron emission tomography (PET)

T

hardware have evolved significantly in the last few years. Dedicated cardiac SPECT scanners have been

RI P

developed utilizing solid-state detectors instead of analog photomultiplier tubes. To address the concerns about radiation dose and long imaging time in conventional SPECT, new collimators and specialized

SC

gantries, which maximize photon sensitivity, have been implemented by vendors. The new SPECT

NU

hardware has been coupled with optimized iterative reconstruction software, which incorporates resolution recovery, and corrections for photon scatter and attenuation, to deliver significant overall

MA

improvements in performance. Since cardiac PET imaging represents a small percentage of cardiac nuclear cardiology imaging, PET hardware developments have been driven primarily by oncological

recovery

ED

markets. All manufacturers have incorporated both time-of-flight (TOF) information and resolution in their reconstructions. Recently, new solid-state photomultipliers to improve detection

PT

sensitivity and resolution have been introduced in PET/ computed tomography (CT), as well as

CE

PET/magnetic resonance (MR) scanners. In this review, we present an update on the latest advances in

SPECT

AC

cardiac SPECT and PET hardware and their relation to the context of cardiovascular (CV) imaging

Recently, several novel designs of the gantry coupled with new solid-state detectors, which allow increased photon sensitivity in the myocardial region, have been introduced. Physical space requirements are reduced since the dedicated detectors and gantries are significantly smaller in comparison to conventional scanners. Due to new reconstruction techniques utilizing resolution recovery principles, these systems are able to maintain or improve overall spatial resolution, and significantly increase photon sensitivity. As a consequence, much shorter acquisition times are possible allowing for the additional benefit of reducing patient motion during scans and increasing patient comfort New Detection systems for SPECT 4

ACCEPTED MANUSCRIPT Radiation detectors made of Cadmium-Zinc-Telluride (CZT) – a compound of cadmium, zinc, and tellurium have been introduced by two vendors for SPECT imaging.

CZT detectors are more expensive than

T

detectors based on NaI(Tl) crystal coupled with photomultiplier tubes in the conventional Anger camera,

RI P

but do have several imaging advantages. The CZT detector’s improved energy response leads to a 30% reduction of the scatter component in measured data [1]. In addition, the solid state crystal has superior

SC

intrinsic spatial resolution as compared to the conventional Anger camera. Moreover, the compact detector size allows for the practical designs of the innovative imaging geometries and high-sensitivity

NU

collimation. A solid state CZT-based detector with associated electronics is shown in Figure 1.

MA

Two vendors have introduced CZT SPECT systems to the market. One design (Spectrum Dynamics DSPECT) uses pixilated CZT detector arrays. These detectors are mounted in 9 vertical columns—4

ED

detectors in each column, and placed in a 90-degree gantry geometry. Each column consists of an array of 1024 CZT elements (2.46 x 2.46 x 5 mm thick), arranged in a 16 x 64 element array with an approximate

PT

size of 40 x 160 mm, allowing for a 16 cm coverage of the thorax. Since CZT crystals are very expensive,

quality and cost.

CE

the vendor recently introduced a system with 6 detector columns, creating a trade-off between image In the other camera design (GE 530c), a curved array of 19 solid-state CZT pixilated

AC

detector ensembles (one for each pinhole collimator) is used with optimized geometry for cardiac imaging and coupled with pinhole collimators (see next section). Each detector ensemble consists of 4 detectors as shown in Figure 1, each of the individual detectors has 246 CZT detector elements (2.46 x 2.46 x 5 mm thick) arranged in a 16 x 16 element array. Solid state detectors can also be used with the scintillation crystal to register light emitted by the scintillation crystal and thus indirectly detect photon signals. This approach is taken by Digirad in their Cardius cameras. In these systems, Cesium Iodide doped with thallium CsI(Tl) scintillation crystals and photodiodes (instead of photomultipliers) are used [4]. This design has been used in a 2- or 3-detector configuration on these systems, allowing for a more compact design than possible with a conventional Anger camera. CsI(Tl) is a better choice than NaI(Tl) when used with silicon photodiodes due to spectral 5

ACCEPTED MANUSCRIPT light response characteristics. In this design, each detector head contains an array of 768 6.1 x 6.1 x 6 mm thick CsI(Tl) crystals, coupled to the individual silicon photodiodes that convert the light output of the

T

crystals to electrical pulses, which are then digitally processed to form an image.

RI P

New SPECT collimators and gantries

SC

The new solid detectors have been combined with innovative collimator designs to increase the photon sensitivity in the myocardial region. Increasing the collimator sensitivity is of primary importance since a

NU

conventional dual head SPECT system detects at most < 0.01% of incoming photons.[2]

MA

One design features 9 or 6 small high-sensitivity Tungsten collimators –one for each of the detector columns (as described above). These collimators can be positioned close to the patient’s body ensuring

ED

optimal imaging geometry (Figure 2). The collimators are square, parallel-hole, and high sensitivity, and the dimensions of the holes (2.26x2.26 mm) with added walls (0.2mm thickness) are matched to the

PT

dimensions of a single detector element. The hole dimensions are much larger than in the conventional high resolution collimators, resulting in much better geometric efficiency, and consequently, much higher

CE

sensitivity at the expense of image resolution. However, the loss in the geometric spatial resolution that

AC

results from this design is compensated by the use of pixelated CZT detectors with superior energy resolution and intrinsic spatial resolution, as well as software-based resolution recovery applied during image reconstruction.[3] Patients are positioned on a chair rather than on a bed, increasing patient comfort; they are typically imaged in supine and upright positions on this camera. An entirely different collimation system and gantry design is used by another CZT camera. The design of this system with pinhole collimation is shown in Figure 3. Pinhole collimation is based on multiple focal depths of 19 pinhole collimators. There is only one hole (4.75 mm in diameter) in each collimator. The CZT detectors and collimators do not move during the acquisition and all photons are acquired simultaneously through all of the pinholes. The use of simultaneously acquired pinhole views allows

6

ACCEPTED MANUSCRIPT consistent tomographic sampling, which reduces motion artifacts. Patients are imaged in the traditional supine and prone positions with their arms placed over their heads.

T

The high sensitivity is achieved by the wider acceptance of photons with multipinhole collimation or

region.

RI P

larger collimator holes in the standard collimator—and limiting the photon detection to the myocardial The high sensitivity characteristics of the new scanners are often mistakenly attributed to the

SC

CZT crystal. Although the density of CZT is higher than NaI(Tl) (5.8 vs. 3.7 g/cm3), due to the energy

NU

resolution and cost considerations, the CZT detectors used in these systems are thinner (~5mm) than typical 3/8 inch (9.5mm) NaI(Tl) crystals, making the overall intrinsic detection efficiency of these

MA

systems comparable to that of the Anger detectors. Therefore, the sensitivity gain is related only to the use of new collimators and dedicated cardiac imaging configurations.

ED

Photon sensitivity can also be increased on conventional Anger cameras by utilizing magnification of the image region of interest with focusing collimator. General conventional dual head cameras can be

PT

equipped with cardio-focal collimators based on a previously developed concept [4]. In such a

CE

converging/diverging collimator, the center of the field-of-view is magnified in axial as well as in transaxial directions, while the diverging collimation (minification) at the periphery of the field-of-view allows

AC

for the coverage of the entire body. This ensures avoidance of truncation artifacts common to fan-beam collimators when imaging the torso. Additionally, the cardio-centric orbit is used to keep the detectors as close to the heart as possible. The number of detected photons from the heart is higher by more than a factor of two, as compared to that of a high resolution parallel-hole collimator [4]. The design of such collimators (Siemens, SmartZoom) is shown in Figure 4.

These collimators can be added as a field-

upgrade to the existing conventional Anger cameras, which can be significantly cheaper than the purchase of a new dedicated CV system.[5] Another SPECT hardware design camera uses traditional fan beam collimators in the 3-detector configuration coupled with solid state (photodiode) detectors and Cs(I) scintillation crystals (Digirad

7

ACCEPTED MANUSCRIPT Cardius 3 XPO). The two outer detectors are positioned at a 67.5 degree angle to the central detectors. Data acquisition is accomplished by rotating the chair (rather than detectors) by 67.5 degrees, resulting in

T

an acquisition arc of 202.5 degrees.

RI P

The new cardiac SPECT scanners described above allow better image resolution and improved count sensitivity as compared to the conventional dual-head Anger camera systems. Recently, Imbert at al.[6]

SC

compared the performance of these new cardiac SPECT systems with the conventional dual head scanner.

NU

In addition, we have obtained additional information describing similar benchmark results from one vendor.[7] The comparisons of count sensitivity (expressed in %), and image resolution (in mm), for all

MA

these new scanners vs. conventional dual head systems are shown in Figure 5. As illustrated, the new cardiac SPECT systems may allow up to a 7-times increase in the sensitivity and over a 2-times

ED

improvement in effective image resolution as compared with the conventional dual head systems. The reported imaging times on these systems range from 2-3 minutes for systems with dedicated solid state

PT

detectors [8] [9] to 4 minutes on the system with dedicated collimators[4].

CE

Attenuation correction for SPECT

AC

CT-based attenuation correction is available on some of the new SPECT scanners with up to a 64-slice CT. The CT component allows for the photon attenuation correction (AC) and also for hybrid imaging incorporating either calcium scoring or CT angiography. These hybrid systems are, however, considerably more expensive than the standalone SPECT units.

Vendors also offer software tools that

allow manual alignment of externally acquired CT scans on separate scanners. Potentially, CT calcium scans (which are obtained with cardiac gating), could also be used for the AC purpose with such manual alignment[10]. To reduce the radiation dose from AC scans, various new designs have been developed for SPECT scanners. One system (Philips, Brightview XCT) uses flat panel x-ray detector system, which can obtain low dose CT-AC images (0.12 mSv). The AC scan of the entire heart volume (14 cm axial field of view) 8

ACCEPTED MANUSCRIPT is acquired in a single 60-second rotation with, while the patient is breathing normally. Therefore, the AC scan is averaged over multiple respiratory cycles and aligns is better aligned with SPECT data.

T

Digirad developed an AC system where photons from an X-ray source are detected by solid state

RI P

detectors with a fan beam collimator operating in a high counting rate mode.[11] The scan time for the acquisition of the AC maps is also 60 seconds (Figure 6). The dose is approximately 0.005mSv. Both of

SC

these systems offer potential advantages (lower cost, good SPECT/CT alignment) as compared to high-

NU

end systems; however, they are not capable of CT angiography imaging or calcium scoring. Some of the new dedicated cardiac cameras may not be available or can be prohibitively expensive in

MA

SPECT/CT configurations. To allow mitigation of attenuation artifacts for the systems without AC capability, novel protocols have been developed with 2 sequential scans in 2 patient positions

ED

(supine/upright or supine/prone–depending on the scanner), allowing for differentiation of true perfusion defects from artifacts for systems without AC[12]. It is conceivable, that the 2-position imaging may also

PT

allow the detection of position-related truncation artifacts, which may occur with the limited field-of-view

CE

gantry of the new scanners [13].

To date, new cardiac SPECT hardware with increased photon sensitivity has been utilized primarily to

AC

shorten the imaging time. However, it is also possible to significantly reduce patient doses [14-18] [19]. The feasibility of standard time, ultra-low dose stress cardiac SPECT with the new hardware has been evaluated in a simulation study, enabled by the list-mode acquisition of SPECT data.[18] To simulate lower injected doses, the original high-dose (21.7 ± 5.4mCi of

99m

Tc injected at stress), full-time (14

minutes), raw list-mode data was sub-sampled. Subsequently, lower-dose equivalent scans were reconstructed retrospectively. The standard error of differences for perfusion quantification computed from the scans with 8 million- and 1 million-counts in the left ventricular (LV) region was similar to the reproducibility of the conventional SPECT [20], providing the evidence that the new high-efficiency SPECT scanners are capable of routine imaging with a radiation dose of < 1 mSv at standard imaging times. 9

ACCEPTED MANUSCRIPT Future developments in SPECT Despite recent impressive improvements in SPECT instrumentation, further hardware improvements are

T

still possible. One area of possible improvement is the additional gain in photon sensitivity, since even

RI P

the latest scanners detect less than 0.1% of the incoming photons. Pre-clinical studies demonstrated that higher efficiency can be achieved in cardiac SPECT than achieved to date with latest hardware. Zeng et

SC

al. [21] has theoretically demonstrated that multi-divergent beam collimators could have twice the

same image resolution.

NU

sensitivity of the currently available multi-pinhole cameras in terms of sensitivity while maintaining the Theoretical simulations of multi-pinhole designs with spherically curved

MA

detectors fitted to pinholes demonstrated potential of close to 50% improvement [22] as compared to the currently available multi-pinhole camera. Nevertheless, these published reports are still only theoretical

effective for clinical applications.

PT

PET

ED

simulations of possible future hardware designs. It remains to be seen if these designs will be cost-

CE

Recent PET/CT systems use fast scintillators lutetium oxyorthosilicate (LSO) or lutetium-yttrium-

AC

orthosilicate (LYSO), which allow shorter coincidence windows and better random events rejection than Bismuth Germanate crystals; they also allow for TOF data collection and image reconstruction. All newer PET/CT scanners operate in 3D mode, without the inter-plane septa. In 3D mode, coincident photons are detected in all directions and not just in slices of inter-plane or cross-planes from adjacent rings, which results in greater photon sensitivity than in 2D mode with inter-plane septa by a factor of 4-6, at the expense of increased scatter fractions from 10-15% to 30-40% [23] (Figure 7). The best reported PET photon sensitivity for 3D PET/CT and PET/MR, and 2D PET/CT systems compared to conventional and high-efficiency SPECT is shown in Figure 8. The 3D PET sensitivity is at least two to three orders of magnitude higher than that of conventional cardiac SPECT due to the lack of physical collimators. Reduced detector element sizes of the new PET scanners allow for high spatial resolution, which range

10

ACCEPTED MANUSCRIPT between 4-5 mm in the center of the field-of-view depending on the model and vendor. The axial coverage of 16-22 cm allows full coverage of the myocardium in one bed position. To improve axial

T

noise uniformity, new designs have been introduced, which can allow continuous movement of the patient

RI P

through the PET scanner, which is similar to the continuous bed motion through a helical CT scan. This technique eliminates overlapping bed acquisitions and maintains uniform noise sensitivity across the

position is required.

SC

entire scan range –it may be applicable to CV rather than cardiac imaging where more than one bed High-count rate performance has been significantly improved in recent PET/CT

NU

systems, allowing for myocardial blood flow measurements with 82Rb tracer[24] .

MA

New PET detection systems

Traditional PET/CT scanners use photomultiplier tubes for photon detection; however, designs featuring

ED

silicon digital photomultipliers (SiPM) [25], or avalanche photodiodes, [26] have recently been introduced. All-digital PET scanners with SiPM detectors have been introduced. LYSO scintillation

PT

crystals are used as in the conventional scanner to convert annihilation photons to the scintillation events.

CE

Subsequently, avalanche photodiodes coupled with low-voltage digital complementary-metal oxide semiconductor circuits combined on the same silicon substrate detect scintillation events. Each SiPM

AC

detector array (32.6x32.6mm) is comprised of 8x8 pixels with active areas of 3.9x3.9mm. In Figure 9, we show both the conventional photomultipliers and SiPM, and their event detection. SiPMs have very good intrinsic timing resolution (44ps). This design, together with TOF methods (see below), allows this system to achieve simultaneous improvements in sensitivity, image resolution, and maximal count rate (important for the myocardial flow measurements with

82

Rb).

Furthermore, SiPM detectors are

insensitive to the electromagnetic interference, which will be critical for the PET/MR configuration. Solid-state photomultipliers of a different kind—avalanche photodiodes—are used by Siemens in their Biograph mMR PET/MR system[27], but not in their PET/CT system. The main reason for the use of avalanche photodiodes in the PET/MR system is their insensitivity to the MR electromagnetic

11

ACCEPTED MANUSCRIPT interference during PET acquisition. A comparison of the performance characteristic of newer 3D

T

PET/CT scanners and the new digital PET/CT are shown in Table 1.

RI P

New PET reconstruction methods and TOF

Current PET/CT systems employ a fully 3D iterative PET reconstruction, which allows incorporation of

SC

the CT attenuation maps and corrections for scatter, attenuation, random events, spatial system response,

gamma-prompt decay.[28]

NU

and dead time. New 3D scatter correction techniques have been developed taking into account the 82Rb In addition to these general improvements in reconstruction, the TOF

MA

reconstruction algorithms have been implemented [29]. Although TOF reconstruction was proposed decades ago, it has only recently become practical due to improved timing resolution achievable by new

ED

coincidence electronics combined with fast scintillators (LSO, LYSO); TOF principle exploits the information provided by the time difference between the arrivals of the 2 annihilation photons at the

PT

opposite detectors during PET detection (Figure 10). [30] This information is used to determine the

CE

improved determination of the positron annihilation location during the iterative reconstruction. TOF allows improved signal-to-noise ratio in PET images, particularly in large subjects.[31]. It has been

AC

shown that TOF imaging results in contrast improvements and increased uptake in the lateral wall for 13Nammonia myocardial perfusion imaging. [32] Most of the latest PET/CT scanners have the TOF option available. An example of improved image contrast and reduced noise obtained with TOF for the cardiac 13

N-ammonia imaging is shown in Figure 11. It is expected that the use of digital SiPMs can further

improve the TOF performance of PET systems as can be seen in Table 1.

A recent improvement to the reconstruction algorithms offered by all PET vendors is the 3D modeling of scanner-specific point spread function maps, which are subsequently used during 3D iterative reconstruction to predict the input signal [29] . This resolution recovery method has been termed by vendors as point spread function or high definition reconstruction. Effective tomographic resolution as 12

ACCEPTED MANUSCRIPT low as 2 mm, and significant improvements in image contrast, have been demonstrated by the application of this method.[33] The combination of the TOF reconstruction and correction of the resolution recovery

T

showed the best improvement of image quality.[34, 35] Clinically, better definition of subtle myocardial

RI P

perfusion defects have been reported with the use of the resolution recovery.[36] The advanced image reconstructions utilized for 3D PET are significantly different from those used previously in 2D PET.

SC

Therefore, the 3D cardiac PET is not equivalent to 2D PET.

NU

The increased photon sensitivity of the new 3D PET/CT systems can be utilized to reduce radiation exposure to the patient. For 82Rb imaging, the typical injected dose for the 2D PET system (operating

MA

with inter-plane septa) has been recommended to be in the range of 40–60 mCi.[37] For 3D PET mode, lower, weight-based injected doses have been proposed—with an average injected dose of 26 mCi—

ED

which translates to 2.4 mSv for the full stress/rest scan and just 1.2 mSv for the stress component.[38] The doses for other perfusion tracers are also relatively low. For example, the radiation dose from a

PT

single 3D PET perfusion scan (stress or rest) is about 1 mSv for 13N-ammonia.[39] A potential benefit of

CE

the lower injected 82Rb dose could be more reliable myocardial perfusion flow data; the high 82Rb count rate during 3D PET acquisition may present challenges in accurate measurements of the LV input

AC

function at higher doses.[40]

Attenuation correction for PET/CT In PET cardiac imaging, AC has been used ever since the introduction of standalone PET with two radionuclide transmission rod-sources circling around the patient to collect photon attenuation maps. The advantages of CT-based AC over the transmission-based AC are low noise attenuation maps, fast data acquisition, and elimination of bias from emission contamination of post-injection transmission scans.[41] The CT disadvantages are higher radiation doses, as well as susceptibility to metal artifacts from pacemakers or implantable cardioverter defibrillators. CT-based AC for PET has proven challenging because of the misalignment between CT and PET images mostly due to respiratory motion [42], which is

13

ACCEPTED MANUSCRIPT especially critical in myocardial perfusion imaging.[43, 44] CT has temporal resolution of less than one second, whereas PET has temporal resolution of about one respiratory cycle. This mismatch has been

T

reported to cause misalignment in 40 to 60% studies.[45-47] It has also been shown that there are

RI P

significant differences between traditional transmission source-based AC (such as Ge-68 or Cs-137) and CT-based AC, which may remain after alignment of CT maps to emission data [44, 48]. Therefore,

SC

careful visual verification of the alignment is needed to minimize registration artifacts. Vendors have developed automatic registration of CT and PET maps to reduce these artifacts[49], but these methods are

NU

not yet routinely applied. Average CT scan, based on cine CT data acquisition over one breath cycle of

MA

about 4 to 5 seconds has been proposed to match the temporal resolution of CT and PET and improve registration of the CT and PET data. Unfortunately, current implementations require respiratory gating devices to generate the average CT images for AC and alignment, which may not be practical for routine

ED

clinical use [50]. It was reported that average CT could reduce the rate and magnitude of mis-registration

of 0.8 mSv.

PT

between CT and PET [44] at an increase of radiation dose by 0.4 mSv [51] over the helical CT protocols Recently, researchers proposed to utilize the CT calcium scans for attenuation

CE

correction.[52] It may be also possible to estimate coronary calcium from non-gated low-dose CT

AC

attenuation maps.[53, 54]

CT component in a hybrid PET/CT or SPECT/CT Typically, the maximum number of slices or detectors for a CT scanner on the hybrid scanners of PET/CT or SPECT/CT is 64, although some vendors are putting out new 128-slice systems, and the largest detector coverage is 4 cm in the axial direction enabling coronary artery CT imaging in about 5 to 6 seconds. Dose efficiency is the percentage of x-ray radiation through the patient and utilized for image reconstruction, and it increases as the detector size increases. For example, on the GE CT scanner, the 64slice and 16-slice are with 4 cm and 2 cm detector coverage and are 95% and 89% dose efficient, respectively. Siemens CT virtually doubles the rows of the detectors by alternating focal spots technology, which increases sampling in the axial direction to enable high resolution CT imaging of 0.3 14

ACCEPTED MANUSCRIPT mm isotropic resolution. Latest PET/CT or SPECT/CT scanners include a 64-slice (or higher) CT component at gantry rotation of 0.33 to 0.35 seconds; therefore, it is possible to combine, in one scanning

T

session, the high-quality coronary CT angiography with myocardial perfusion PET.[55, 56] Nevertheless,

RI P

cardiac PET and coronary CT angiography images obtained on a hybrid PET/CT scanner are not truly simultaneous since they are obtained in different breathing patterns. This may result in significant

SC

misalignments between PET and CT data and will require software registration.[57] 16-slice CT is sufficient for hybrid PET/CT or SPECT/CT to incorporate the CT calcium scan [58] into routine hybrid

NU

clinical imaging.

MA

Respiratory gating

Image blurring and CT-AC misregistration artifacts caused by the movement of the heart during breathing

ED

could be potentially corrected by respiratory gating.[59, 60] Breathing motion can be detected by an infrared camera, registering the movement of an infrared reflective marker on the patient's abdomen,

PT

either by an inductive respiration monitor with an elastic bellows or pressure sensor around the patient’s

CE

chest, or by measurements of patient respiration (spirometry). [61-63] These external signals are used as triggers to assign the PET events into appropriate respiratory phases during image reconstruction.

AC

Respiratory gating can also be used for correction of the upward creep of the heart during exercise-stress imaging (as the heart-rate slows down and the body relaxes over time). PET/CT scanners can be equipped with additional hardware for respiratory gating by the vendors. Simultaneous acquisition of both respiratory and cardiac gating signals using bio-impedance techniques and signals obtained from standard ECG electrodes during cardiac PET [64] and techniques based on accelerometers [65] have been recently proposed. These techniques have the potential to simplify the challenging logistics of dual gating for cardiac imaging. In addition self-gated methods that allow for the derivation of respiratory-gating signal retrospectively from the raw list-mode PET data rather than using an external device (self-gating or datadriven gating) have been developed. [60] Nevertheless, respiratory gating is not used in routine cardiac PET or SPECT imaging, due to the need for external hardware and complexity of acquisition. 15

ACCEPTED MANUSCRIPT PET/MR systems Hybrid PET/MR has been recently introduced by at least three vendors for clinical use. At least 2 vendors

T

offer systems that are capable of collecting MR and PET image data simultaneously with almost no

RI P

compromise of either MR or PET functionality and performance [27]. The simultaneous acquisition of MR and PE data presents significant engineering challenges due to cross-interference of PET and MR

SC

signals. To minimize this cross-reference in the PET component, analogue photomultipliers are replaced

NU

with high-speed solid-state avalanche photodiode detectors (APDs) in (Siemens mMR) or SiPMs (GE Signa), which are insensitive to magnetic fields.[66] The Siemens MR system is based on the Verio MR

MA

scanner with a modified body radiofrequency coil, a gradient coil, and a patient handling system. The PET detector system is arrayed around a cylinder that is inserted into the bore of the MR scanner and thus

ED

reduces the diameter of the bore from 70 to 60 cm [67] . LSO scintillator crystals (4 x 4 x 20 mm), combined into blocks (8 x 8 crystals), are in turn combined into a cylinder of detectors (8 axial x 56

PT

circumferential blocks). Each detector block consists of 64 crystals, a 3x3 array of silicon APDs, and

CE

electronics for amplification and pulse shaping. The MR gradient system is cooled and the PET detectors are shielded with copper in order to reduce electromagnetic interference. The APDs do not distort the

AC

main magnetic field of the MR, and the acquisitions of the MR and PET systems are synchronized to reduce any possible distortion of the detected PET impulses by the MR radiofrequency and gradient pulses. The PET component of the Siemens scanner has 25.8 cm axial field of view and improved photon sensitivity of 1.5% as compared to 0.95% sensitivity of their PET/CT system.[27] However, the mMR system does not offer the TOF reconstruction, due to relatively low timing resolution of APDs. The Signa PET/MR, on the other hand, uses SiPM to achieve a TOF with coincidence timing resolution of 400 ps and offers a 25 cm PET axial field of view to achieve photon sensitivity of 2.1%. The static magnetic field can reduce the effective positron range and potentially improve the PET image resolution. This could potentially be helpful in

82

Rb cardiac imaging since this tracer has much higher average kinetic

16

ACCEPTED MANUSCRIPT energy of the positrons, but will not be important for 18F-based tracers imaging due to the much shorter

T

positron range.

applications have been reported.[68-71]

RI P

The primary application for PET/MR imaging is currently in oncological imaging, but several clinical CV In comparison to PET/CT, PET/MR can reduce the overall

SC

patient’s radiation dose. In addition, advanced functional and molecular MR techniques not available for CT could be of value for cardiac imaging. The main PET/MR disadvantage, as compared to PET/CT, is

NU

the significantly higher cost. Furthermore, the MR electromagnetic field may prevent imaging patients with pacemakers, implantable cadioverter defibrillators, or other implanted devices susceptible to

MA

electromagnetic interference. It is likely that PET/MR will be exploited as a powerful research tool for

ED

direct cross-comparison of new PET or MR cardiac imaging techniques.

PT

Attenuation correction for PET/MR

MR images do not contain information about photon attenuation, and therefore, in PET/MR systems the

CE

attenuation maps must be estimated indirectly. Software methods have been proposed for MR-based

AC

attenuation correction, based on image segmentation, or coregistration of templates with pre-assigned coefficients. An exact attenuation correction is crucial in cardiac PET imaging. A method proposed by Martinez-Moeller et al. [72] has been utilized in clinical systems, with a two-point Dixon VIBE MR sequence to obtain fat and water compartments for calculating attenuation. Lung tissue in the MR images is identified by image segmentation based on symmetry principles and relative bed position. Softwareconstructed attenuation maps obtained from MR may not reflect true photon attenuation, since only few tissue classes can be defined. In addition, the actual attenuation coefficients may be different than standard assumed values—especially in the lungs.[68]

Nevertheless, the simultaneous PET/MR

acquisition offers some advantage as compared to PET/CT attenuation correction due to simultaneous

17

ACCEPTED MANUSCRIPT acquisition and exact registration. Attenuation correction with MR can be applicable to cardiac region imaging as reported in several studies [73-75].

RI P

T

Conclusions

Advances in cardiac SPECT hardware and associated acquisition software allow for much higher photon

SC

sensitivity and improved image resolution compared to traditional Anger cameras. Vendors also offer dedicated high-efficiency cardiac collimation systems, which can be retrofitted on conventional scanners

NU

–reducing the cost required for equipment upgrades. Some of the new SPECT systems are combined with CT scanners, which allow for attenuation correction. However, to reduce the scanner cost and radiation

MA

dose, innovative devices have been developed for attenuation correction. Furthermore, 2-position imaging, which can mitigate the imaging artifacts without the need for additional hardware, is facilitated

ED

by the high sensitivity of the new SPECT scanners. Recent PET instrumentation improvements include high sensitivity 3D PET and new reconstruction methods. 3D PET without inter-plane septa has become

PT

the primary PET/CT configuration, allowing reduced radiation dose and imaging time. Resolution

CE

recovery methods combined with TOF are now implemented by all vendors, allowing improved contrast, enhanced image resolution, and reduced image noise .Vendors introduced solid state PET detectors

AC

featuring digital silicon photomultipliers or avalanche photodiodes. These new detectors offer improved imaging characteristics (higher count rate and image resolution) and can operate in electromagnetic MR fields—allowing simultaneous PET/MR imaging with the new hybrid PET/MR scanners. The higher maximum count rate of the new PET scanners will allow routine first-pass 82Rb imaging with 3D PET. The attenuation correction on the new PET/MR scanners requires improved software methods, but carries the potential to remove errors associated with misregistration of PET/CT maps, which are common in cardiac PET/CT.

18

ACCEPTED MANUSCRIPT

Table 1.

TF

OpenView

70

Patient scan range [cm]

190

200

Maximum patient weight [kg (lb)]

195 (430)

226 (500)

Plane spacing [mm]

2 or 4

NU

45 or 90

78

70

195

190

226 (500)

195 (430)

3D S&S, continuous

3D S&S

47

109

72

3.27

2

1, 2, or 4

4.2 x 6.3 x 25

4 x 4 x 20

4 x 4 x 22

28,336

13,824

32,448

23,040

420

256

768

23,040 SiPM's

18

15.7

21.8

16.3

LYSO

LYSO

LSO

LYSO

System sensitivity 3D, [%]*

0.74

0.75

0.95

> 1.0

Trans axial resolution @ 1 cm [mm]*

4.7

4.9

4.4

4.0

Trans axial resolution @ 10 cm [mm]*

5.2

5.5

4.9

4.5

Axial resolution @ 1 cm [mm]*

4.7

5.6

4.5

4.0

Axial resolution @ 10 cm,[mm]*

5.2

6.3

5.9

4.5

120

130

175

400

Peak NECR [kcps]

@19 kBq/ml

@29.5 kBq/ml

@28 kBq/ml

@30 kBq/ml

Time-of-Flight resolution [picoseconds]

591

544

540

307

ED

4 x 4 x 22

PT

Crystal size [mm]

Detector material

AC

Physical axial FOV [cm]

CE

Number of crystals Number of PMTs

3D S&S

MA

Number of image planes

SC

Patient port [cm]

3D S&S

Vereos

Flow

70

Acquisition modes

Biograph mCT

RI P

Model Product Name

Discovery 710

T

Ingenuity

19

ACCEPTED MANUSCRIPT

8.9

8.2

8.1

4.6

Coincidence window [nanoseconds]

4.5

4.9

4.1

1.5

*NEMA 2001

S&S: Step and shoot

RI P

T

Time-of-Flight localization [cm]

AC

CE

PT

ED

MA

NU

SC

Table 1. Comparison of the Philips Ingenuity TF, GE Discovery 710, Biograph mCT Flow and the new Philips digital PET/CT Vereos. The sensitivity, NECR (noise equivalent count rate), coincidence window and TOF resolution are favorable toward Philips digital PET/CT. FOV–field-of-view, PMT– photomultiplier tubes, NECR–noise equivalent count rate, kcps–kilocounts per second, kBq/ml – kiloBecquerel/milliliter.

20

ACCEPTED MANUSCRIPT

Figure legends

T

Figure 1. Typical configuration of a single digital detector coupled with the CZT crystal. (ASIC-

RI P

application specific integrated circuit).

Figure 2. Acquisition geometry of the D-SPECT system with 9 detector columns (A) and 6 detector

SC

columns (B). One detector column (C).

NU

Figure 3. Acquisition geometry of the GE NM530c multipinhole system (A), multipinhole collimator (B), and imaging formation diagram at each pinhole (C) (DH –detected height, SH- subject height, DPD –

MA

detector to pinhole distance, SPD –subject to pinhole distance).

ED

Figure 4. Siemens IQ∙SPECT acquisition geometry with the diverging/converging (cardiofocal) collimator (A) allowing increased sensitivity in the heart region by selective focusing in the maximum

PT

zoom area. Projection images obtained with conventional (B) and cardiofocal (C) collimators.

CE

Figure 5. Absolute count sensitivity in % (left axis) and tomographic image resolution in mm (right axis) of the new cardiac SPECT systems as compared to the conventional dual head system. DSPECT -

AC

Spectrum Dynamics DSPECT system; NM – GE Alcyone 530c/570c system; IQSPECT – Symbia system with dual head astigmatic collimators from Siemens; Cardius Digirad X∙ACT with fan beam (FB27) collimator and n-speed reconstruction. Conventional= conventional dual headed SPECT with low energy high resolution collimators. Data based on Imbert et al. [6] phantom comparison and on specifications provided by Digirad.[7] .* Digirad XACT smallest orbit radius is approximately 20 cm and testing was conducted at that distance, as compared to standard 15 cm for NEMA NU-1 2001. Figure 6. Low-dose attenuation correction system for the upright dedicated cardiac system (Cardius, Digirad). Adapted with permission from Conwell et al. [11]

21

ACCEPTED MANUSCRIPT Figure 7. The principle of 2D (left) and 3D (right) PET acquisition. In 2D mode, the interplane septa are reducing the number of coincident photons, but are also reducing the scatter.

T

Figure 8. Comparison of highest absolute photon sensitivities reported different nuclear medicine

RI P

modalities: 3D PET/MR, 3D PET/CT, 2D PET/CT, fast SPECT and conventional dual head system. Based on data from Imbert et al. [6], Jakoby et al. [29], Mawlawi et al.[76], and manufacturers’

SC

specifications.

NU

Figure 9. Comparison of analog and digital photon counting with digital silicon photomultipliers. Digital counting results in higher spatial resolution and faster timing resolution. Images courtesy Philips

MA

Healthcare.

ED

Figure 10. Principle of time-of flight (TOF) imaging in PET. Conventional reconstruction back-projects data with equal weights throughout the image space (A), TOF reconstruction with LSO/LYSO detector

PT

photo-multiplier tubes (PMT) back-projects data with Gaussian weights at the most likely coincidence location (B), and next generation TOF reconstruction with digital silicon multipliers (SiPM) further

CE

narrows the range of coincidence location to improve reconstruction accuracy (C).

AC

Figure 11. Comparison of images between standard 3D iterative reconstruction (top) and TOF reconstruction (bottom) for N-13 PET imaging in a case with three-vessel coronary artery disease. There is a severe decrease of uptake in the inferior wall more pronounced with TOF. Reproduced with permission from Timiyama et al. [77]

22

ACCEPTED MANUSCRIPT

REFERENCES

7. 8.

9.

10.

11.

12.

13. 14. 15. 16. 17.

18.

T

RI P

SC

NU

6.

MA

5.

ED

4.

PT

3.

CE

2.

Volokh, L., et al., Effect of detector energy response on image quality of myocardial perfusion SPECT. Nuclear Science Symposium Conference Record, 2008. NSS'08. IEEE, 2008: p. 4043-4046. Gunter, D.L., Gamma Camera Collimator Characteristics and Design, in Nuclear Medicine, R.E. Henkin, Editor. 2006, Elsevier: Philadelphia, PA. p. 107-126. Sharir T, et al., High-speed myocardial perfusion imaging Initial clinical comparison with conventional dual detector anger camera imaging. J Am Coll Cardiol Cardiovasc Imaging 2008. 1: p. 156-63. Hawman, P.C. and E.J. Haines, The cardiofocal collimator: a variable-focus collimator for cardiac SPECT. Phys Med Biol, 1994. 39(3): p. 439-50. Vija, H., J. Chapman, and M. Ray, IQ•SPECT technology White Paper. 2008, Siemens Medical Solutions USA Molecular Imaging. p. 1-7. Imbert, L., et al., Compared Performance of High-Sensitivity Cameras Dedicated to Myocardial Perfusion SPECT: A Comprehensive Analysis of Phantom and Human Images. Journal of Nuclear Medicine, 2012. 53(12): p. 1897-1903. Conwell, R., Digirad system performance (personal communication), P. Slomka, Editor. 2013: Los Angeles. Buechel, R.R., et al., Ultrafast nuclear myocardial perfusion imaging on a new gamma camera with semiconductor detector technique: first clinical validation. European journal of nuclear medicine and molecular imaging, 2010. 37(4): p. 773-778. Sharir, T., et al., Multicenter trial of high-speed versus conventional single-photon emission computed tomography imaging: quantitative results of myocardial perfusion and left ventricular function. J Am Coll Cardiol, 2010. 55(18): p. 1965-74. Schepis, T., et al., Use of coronary calcium score scans from stand-alone multislice computed tomography for attenuation correction of myocardial perfusion SPECT. European journal of nuclear medicine and molecular imaging, 2007. 34(1): p. 11-19. Conwell, R., et al., Phantom evaluation of a cardiac SPECT/VCT system that uses a common set of solid-state detectors for both emission and transmission scans. Journal of Nuclear Cardiology, 2010. 17(3): p. 459-469. Nakazato, R., et al., Quantitative upright-supine high-speed SPECT myocardial perfusion imaging for detection of coronary artery disease: correlation with invasive coronary angiography. J Nucl Med, 2010. 51(11): p. 1724-31. Fiechter, M., et al., Cadmium-zinc-telluride myocardial perfusion imaging in obese patients. Journal of Nuclear Medicine, 2012. 53(9): p. 1401-1406. Henzlova, M. and W. Duvall, The future of SPECT MPI: Time and dose reduction. Journal of Nuclear Cardiology, 2011. 18(4): p. 580-587. Gimelli, A., et al., High diagnostic accuracy of low-dose gated-SPECT with solid-state ultrafast detectors: preliminary clinical results. Eur J Nucl Med Mol Imaging, 2012. 39: p. 83-90. Nkoulou, R., et al., Semiconductor detectors allow low-dose-low-dose 1-day SPECT myocardial perfusion imaging. J Nucl Med, 2011. 52(8): p. 1204-9. Bateman, T., et al., Prospective Study of Ultra-Low Dose Stress-Only Solid-State SPECT: Comparison of Efficiency, Dosimetry and Outcomes Versus Traditional-Dose AttenuationCorrected Stress-Only Anger SPECT. J Am Coll Cardiol, 2012. 59: p. E1316 (abstract). Nakazato, R., et al., Myocardial perfusion imaging with a solid state camera: Simulation of a very low dose imaging protocol. J Nucl Med, 2013. 54(3): p. 373-9.

AC

1.

23

ACCEPTED MANUSCRIPT

26. 27. 28. 29. 30. 31. 32.

33. 34. 35.

36. 37. 38.

39. 40.

T

RI P

SC

NU

25.

MA

23. 24.

ED

22.

PT

21.

CE

20.

Duvall, W., et al., Reduced isotope dose with rapid SPECT MPI imaging: Initial experience with a CZT SPECT camera. Journal of Nuclear Cardiology, 2010. 17(6): p. 1009-1014. Xu, Y., et al., Automatic and visual reproducibility of perfusion and function measures for myocardial perfusion SPECT. Journal of Nuclear Cardiology, 2010. 17(6): p. 1050-7. Zeng, G.L. and A.M. Stevens, Multidivergent-beam stationary cardiac SPECT. Medical Physics, 2009. 36: p. 2860-2869. Dey, J., Improvement of Performance of Cardiac SPECT Camera Using Curved Detectors With Pinholes. Nuclear Science, IEEE Transactions on, 2012. 59(2): p. 334-347. Fahey, F.H., Data acquisition in PET imaging. J Nucl Med Technol, 2002. 30(2): p. 39-49. Klein, R., R. Beanlands, and R. deKemp, Quantification of myocardial blood flow and flow reserve: Technical aspects. Journal of Nuclear Cardiology, 2010. 17(4): p. 555-570. van Dam, H.T., S. Seifert, and D.R. Schaart, The statistical distribution of the number of counted scintillation photons in digital silicon photomultipliers: model and validation. Phys Med Biol, 2012. 57(15): p. 4885-903. Pichler, B., et al., Studies with a prototype high resolution PET scanner based on LSO-APD modules. Nuclear Science, IEEE Transactions on, 1998. 45(3): p. 1298-1302. Delso, G., et al., Performance measurements of the Siemens mMR integrated whole-body PET/MR scanner. Journal of Nuclear Medicine, 2011. 52(12): p. 1914-1922. Esteves, F., et al., Prompt-gamma compensation in Rb-82 myocardial perfusion 3D PET/CT. Journal of Nuclear Cardiology, 2010. 17(2): p. 247-253. Jakoby, B.W., et al., Physical and clinical performance of the mCT time-of-flight PET/CT scanner. Phys Med Biol, 2011. 56(8): p. 2375-89. Sitek, A., Data analysis in emission tomography using emission-count posteriors. Physics in medicine and biology, 2012. 57(21): p. 6779. Lois, C., et al., An assessment of the impact of incorporating time-of-flight information into clinical PET/CT imaging. Journal of Nuclear Medicine, 2010. 51(2): p. 237-245. Tomiyama, T., et al., Impact of time-of-flight on qualitative and quantitative analyses of myocardial perfusion PET studies using 13N-ammonia. Journal of Nuclear Cardiology, 2014: p. 110. Panin, V.Y., et al., Fully 3-D PET reconstruction with system matrix derived from point source measurements. IEEE Trans Med Imaging., 2006. 25(7): p. 907-21. Bettinardi, V., et al., Physical performance of the new hybrid PET∕ CT Discovery-690. Medical physics, 2011. 38(10): p. 5394-5411. Akamatsu, G., et al., Improvement in PET/CT Image Quality with a Combination of Point-Spread Function and Time-of-Flight in Relation to Reconstruction Parameters. Journal of Nuclear Medicine, 2012. 53(11): p. 1716-22. Le Meunier, L., et al., Enhanced definition PET for cardiac imaging. Journal of Nuclear Cardiology, 2010. 17(3): p. 414-426. Di Carli, M.F., et al., Clinical Myocardial Perfusion PET/CT. J Nucl Med, 2007. 48(5): p. 783-793. Kaster, T., et al., Accuracy of low-dose rubidium-82 myocardial perfusion imaging for detection of coronary artery disease using 3D PET and normal database interpretation. Journal of Nuclear Cardiology, 2012. 19(6): p. 1135-1145. Knuuti, J., Integrated positron emission tomography/computed tomography (PET/CT) in coronary disease. Heart, 2009. 95(17): p. 1457-1463. Tout, D., et al., Assessment of a protocol for routine simultaneous myocardial blood flow measurement and standard myocardial perfusion imaging with rubidium-82 on a high count rate positron emission tomography system. Nuclear medicine communications, 2012. 33: p. 1202-11.

AC

19.

24

ACCEPTED MANUSCRIPT

47.

48. 49. 50. 51. 52.

53.

54.

55. 56.

57.

58. 59.

T

RI P

SC

NU

46.

MA

45.

ED

44.

PT

43.

CE

42.

Zaidi, H., Is radionuclide transmission scanning obsolete for dual-modality PET/CT systems? European journal of nuclear medicine and molecular imaging, 2007. 34(6): p. 815-818. Pan, T., et al., Attenuation correction of PET images with respiration-averaged CT images in PET/CT. J Nucl Med, 2005. 46(9): p. 1481-7. Bacharach, S.L., PET/CT Attenuation Correction: Breathing Lessons. J Nucl Med, 2007. 48(5): p. 677-9. Gould, K.L., et al., Frequent diagnostic errors in cardiac PET/CT due to misregistration of CT attenuation and emission PET images: a definitive analysis of causes, consequences, and corrections. J Nucl Med, 2007. 48(7): p. 1112-21. Gould, K.L., et al., Frequent diagnostic errors in cardiac PET/CT due to misregistration of CT attenuation and emission PET images: A definitive analysis of causes, consequences, and corrections. Journal of Nuclear Medicine, 2007. 48(7): p. 1112-1121. Klingensmith, W.C., 3rd, et al., Decreased perfusion in the lateral wall of the left ventricle in PET/CT studies with 13N-ammonia: evaluation in healthy adults. J Nucl Med Technol, 2009. 37(4): p. 215-9. Slomka, P.J., et al., Automatic registration of misaligned CT attenuation correction maps in Rb-82 PET/CT improves detection of angiographically significant coronary artery disease. J Nucl Cardiol (in press), 2015. Slomka, P.J., et al., Comparison of myocardial perfusion 82Rb PET performed with CT- and transmission CT-based attenuation correction. J Nucl Med, 2008. 49(12): p. 1992-8. Bond, S., et al. Automatic registration of cardiac PET/CT for attenuation correction. in Nuclear Science Symposium Conference Record, 2008. NSS'08. IEEE. 2008. Dresden, Germany: IEEE. Alessio, A.M., et al., Cine CT for attenuation correction in cardiac PET/CT. Journal of Nuclear Medicine, 2007. 48(5): p. 794-801. Pan, T., et al., Attenuation correction of PET cardiac data with low-dose average CT in PET/CT. Med Phys, 2006. 33(10): p. 3931-8. Burkhard, N., et al., Coronary calcium score scans for attenuation correction of quantitative PET/CT 13 N-ammonia myocardial perfusion imaging. European journal of nuclear medicine and molecular imaging, 2010. 37(3): p. 517-521. Einstein, A.J., et al., Agreement of visual estimation of coronary artery calcium from low-dose CT attenuation correction scans in hybrid PET/CT and SPECT/CT with standard Agatston score. Journal of the American College of Cardiology, 2010. 56(23): p. 1914-1921. Mylonas, I., et al., Measuring coronary artery calcification using positron emission tomographycomputed tomography attenuation correction images. European Heart Journal–Cardiovascular Imaging, 2012. 13(9): p. 786-792. Di Carli, M.F., S. Dorbala, and R. Hachamovitch, Integrated cardiac PET-CT for the diagnosis and management of CAD. Journal of Nuclear Cardiology, 2006. 13(2): p. 139-144. Danad, I., et al., Hybrid imaging using quantitative H215O PET and CT-based coronary angiography for the detection of coronary artery disease. Journal of Nuclear Medicine, 2013. 54(1): p. 55-63. Slomka, P., et al., Automatic registration of misaligned CT attenuation correction maps in Rb-82 PET/CT improves detection of angiographically significant coronary artery disease. J Nucl Cardiol, 2015. (in press). Hong, C., et al., Coronary Artery Calcium: Absolute Quantification in Nonenhanced and Contrastenhanced Multi–Detector Row CT Studies1. Radiology, 2002. 223(2): p. 474-480. Martinez-Möller, A., et al., Dual cardiac–respiratory gated PET: implementation and results from a feasibility study. European journal of nuclear medicine and molecular imaging, 2007. 34(9): p. 1447-1454.

AC

41.

25

ACCEPTED MANUSCRIPT

67.

68. 69. 70. 71. 72.

73.

74.

75. 76. 77.

T

RI P

SC

66.

NU

65.

MA

64.

ED

63.

PT

62.

CE

61.

Büther, F., et al., List mode–driven cardiac and respiratory gating in pet. Journal of Nuclear Medicine, 2009. 50(5): p. 674-681. Li, X.A., C. Stepaniak, and E. Gore, Technical and dosimetric aspects of respiratory gating using a pressure-sensor motion monitoring system. Medical physics, 2006. 33(1): p. 145-154. Nehmeh, S., et al., Quantitation of respiratory motion during 4D-PET/CT acquisition. Medical physics, 2004. 31(6): p. 1333-1338. Zhang, T., et al., Application of the spirometer in respiratory gated radiotherapy. Medical physics, 2003. 30(12): p. 3165-3171. Koivumaki, T., et al., An integrated bioimpedance--ECG gating technique for respiratory and cardiac motion compensation in cardiac PET. Phys Med Biol, 2014. 59(21): p. 6373-85. Jafari Tadi, M., et al., Accelerometer-Based Method for Extracting Respiratory and Cardiac Gating Information for Dual Gating during Nuclear Medicine Imaging. Int J Biomed Imaging, 2014. 2014: p. 690124. Judenhofer, M.S., et al., Simultaneous PET-MRI: a new approach for functional and morphological imaging. Nature medicine, 2008. 14(4): p. 459-465. Quick HH, R. Ladebeck, and J. Georgi Whole-Body MR/PET Hybrid Imaging: Technical Considerations, Clinical Workflow, and Initial Results. MAGNETOM Flash 1/2011 ·, 2011. 1, 88100. Rischpler, C., et al., Hybrid PET/MR imaging of the heart: potential, initial experiences, and future prospects. Journal of Nuclear Medicine, 2013. 54(3): p. 402-415. Ratib, O., R. Nkoulou, and M. Schwaiger, Cardiovascular clinical applications of PET/MRI. Clinical and Translational Imaging, 2013: p. 1-7. Quick, H.H., Integrated PET/MR. J Magn Reson Imaging, 2014. 39(2): p. 243-58. Adenaw, N. and M. Salerno, PET/MRI: Current state of the art and future potential for cardiovascular applications. Journal of Nuclear Cardiology, 2013. 20(6): p. 976-989. Martinez-Möller, A., et al., Tissue classification as a potential approach for attenuation correction in whole-body PET/MRI: evaluation with PET/CT data. Journal of Nuclear Medicine, 2009. 50(4): p. 520-526. Kershah, S., et al., Comparison of Standardized Uptake Values in Normal Structures Between PET/CT and PET/MRI in an Oncology Patient Population. Molecular Imaging and Biology, 2013: p. 1-10. Marshall, H.R., et al., Description and assessment of a registration-based approach to include bones for attenuation correction of whole-body PET/MRI. Medical physics, 2013. 40(8): p. 082509. Lau, J.M., et al., Feasibility of MRI attenuation correction in cardiac FDG-PET. Journal of Cardiovascular Magnetic Resonance, 2013. 15(Suppl 1): p. O61. Mawlawi, O., et al., Performance characteristics of a newly developed PET/CT scanner using NEMA standards in 2D and 3D modes. Journal of Nuclear Medicine, 2004. 45(10): p. 1734-1742. Tomiyama, T., et al., Impact of time-of-flight PET in a quantitative analysis on myocardial perfusion studies using 13N-ammonia. Journal of Nuclear Medicine, 2014. 55(supplement 1): p. 1742-1742.

AC

60.

26

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Fig. 1

27

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Fig. 2

28

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Fig. 3

29

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Fig. 4

30

0

SC

0 0

NU

0

MA

0 0

ED

0 0

12

9

6

0

CE

0

15

3

PT

0

AC

Sensitivity [%]

0

18

Fig. 5

31

Resolution [mm]

RI P

T

ACCEPTED MANUSCRIPT

Fig. 6

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

32

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

MA

2D

3D

Fig. 7

AC

CE

PT

ED

2D mode (with septa) acquires coincidence photons in the interplanes and cross-planes and 3D mode acquires data from all angles

33

ED

MA

NU

Modality

SC

RI P

T

ACCEPTED MANUSCRIPT

0.2 0.6 1 1.4 1.8 2.2 2.6 3 3.4 3.8 4.2 4.6 5 5.4 5.8 6.2 6.6 7 7.4 7.8 8.2 8.6 9 9.4 9.8 0 0.4 0.8 1.2 1.6 2 2.4 2.8 3.2 3.6 4 4.4 4.8 5.2 5.6 6 6.4 6.8 7.2 7.6 8 8.4 8.8 9.2 9.6 10

Fig. 8

AC

CE

PT

Absolute photon sensitivity [%]

34

Fig. 9

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

35

Fig. 10

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

36

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Fig. 11

37