Felipe de Galiza Barbosa

Workflow in Simultaneous PET/MRI

The advent of simultaneous PET/MRI brought a large amount of possibilities in research and clinical applications into hybrid imaging. Unlike in PET/CT protocols, the MR component provides an almost unlimited number of pulse sequences and possibilities of different protocols in PET/MRI. Nevertheless, there is an imperative to reduce excessive imaging protocols to realistic clinical practice imaging acquisition. The design of a concise and indication-adapted protocol that provides an efficient workflow in a clinical reality is necessary to transform PET/MRI to a cost-effective imaging modality in addition to PET/CT. The aim of the current article is to point out the main considerations regarding workflow, imaging protocols, and image analysis in simultaneous PET/MRI system in oncology and share our thoughts and experience in acquisition optimization compared with the current literature.

Dose Optimization in TOF-PET/MR Compared to TOF-PET/CT

Purpose To evaluate the possible activity reduction in FDG-imaging in a Time-of-Flight (TOF) PET/MR, based on cross-evaluation of patient-based NECR (noise equivalent count rate) measurements in PET/CT, cross referencing with phantom-based NECR curves as well as initial evaluation of TOF-PET/MR with reduced activity. Materials and Methods A total of 75 consecutive patients were evaluated in this study. PET/CT imaging was performed on a PET/CT (time-of-flight (TOF) Discovery D 690 PET/CT). Initial PET/MR imaging was performed on a newly available simultaneous TOF-PET/MR (Signa PET/MR). An optimal NECR for diagnostic purposes was defined in clinical patients (NECRP) in PET/CT. Subsequent optimal activity concentration at the acquisition time ([A]0) and target NECR (NECRT) were obtained. These data were used to predict the theoretical FDG activity requirement of the new TOF-PET/MR system. Twenty-five initial patients were acquired with (retrospectively reconstructed) different imaging times equivalent for different activities on the simultaneous PET/MR for the evaluation of clinically realistic FDG-activities. Results The obtained values for NECRP, [A]0 and NECRT were 114.6 (± 14.2) kcps (Kilocounts per second), 4.0 (± 0.7) kBq/mL and 45 kcps, respectively. Evaluating the NECRT together with the phantom curve of the TOF-PET/MR device, the theoretical optimal activity concentration was found to be approximately 1.3 kBq/mL, which represents 35% of the activity concentration required by the TOF-PET/CT. Initial evaluation on patients in the simultaneous TOF-PET/MR shows clinically realistic activities of 1.8 kBq/mL, which represent 44% of the required activity. Conclusion The new TOF-PET/MR device requires significantly less activity to generate PET-images with good-to-excellent image quality, due to improvements in detector geometry and detector technologies. The theoretically achievable dose reduction accounts for up to 65% but cannot be fully translated into clinical routine based on the coils within the FOV and MR-sequences applied at the same time. The clinically realistic reduction in activity is slightly more than 50%. Further studies in a larger number of patients are needed to confirm our findings.

Clinical Evaluation of PET Image Quality as a Function of Acquisition Time in a New TOF-PET/MRI Compared to TOF-PET/CT—Initial Results

PurposeThe purpose of this study was to compare only the performance of the PET component between a TOF-PET/CT (henceforth noted as PET/CT) scanner and an integrated TOF-PET/MRI (henceforth noted as PET/MRI) scanner concerning image quality parameters and quantification in terms of standardized uptake value (SUV) as a function of acquisition time (a surrogate of dose). The CT and MR image quality were not assessed, and that is beyond the scope of this study.ProceduresFive brain and five whole-body patients were included in the study. The PET/CT scan was used as a reference and the PET/MRI acquisition time was consecutively adjusted, taking into account the decay between the scans in order to expose both systems to the same amount of the emitted signal. The acquisition times were then retrospectively reduced to assess the performance of the PET/MRI for lower count rates. Image quality, image sharpness, artifacts, and noise were evaluated. SUV measurements were taken in the liver and in the white matter to compare quantification.ResultsQuantitative evaluation showed strong correlation between PET/CT and PET/MRI brain SUVs. Liver correlation was good, however, with lower uptake estimation in PET/MRI, partially justified by bio-redistribution. The clinical evaluation showed that PET/MRI offers higher image quality and sharpness with lower levels of noise and artifacts compared to PET/CT with reduced acquisition times for whole-body scans while for brain scans there is no significant difference.ConclusionThe TOF-PET/MRI showed higher image quality compared to TOF-PET/CT as tested with reduced imaging times. However, this result accounts mainly for body imaging, while no significant differences were found in brain imaging.

Effect of Time-of-Flight Information on PET/MR Reconstruction Artifacts: Comparison of Free-breathing versus Breath-hold MR-based Attenuation Correction.

Purpose To evaluate the magnitude and anatomic extent of the artifacts introduced on positron emission tomographic (PET)/magnetic resonance (MR) images by respiratory state mismatch in the attenuation map. Materials and Methods The method was tested on 14 patients referred for an oncologic examination who underwent PET/MR imaging. The acquisition included standard PET and MR series for each patient, and an additional attenuation correction series was acquired by using breath hold. PET data were reconstructed with and without time-of-flight (TOF) information, first by using the standard free-breathing attenuation map and then again by using the additional breath-hold map. Two-tailed paired t testing and linear regression with 0 intercept was performed on TOF versus non-TOF and free-breathing versus breath-hold data for all detected lesions. Results Fluorodeoxyglucose-avid lesions were found in eight of the 14 patients included in the study. The uptake differences (maximum standardized uptake values) between PET reconstructions with free-breathing versus breath-hold attenuation ranged, for non-TOF reconstructions, from -18% to 26%. The corresponding TOF reconstructions yielded differences from -15% to 18%. Conclusion TOF information was shown to reduce the artifacts caused at PET/MR by respiratory mismatch between emission and attenuation data.

PET/MR Outperforms PET/CT in Suspected Occult Tumors.

Background To compare the diagnostic accuracy of PET/MR and PET/CT in patients with suspected occult primary tumors. Methods This prospective study was approved by the institutional review board. Sequential PET/CT-MR was performed in 43 patients (22 male subjects; median age, 58 years; range, 20-86 years) referred for suspected occult primary tumors. Patients were assessed with PET/CT and PET/MR for the presence of a primary tumor, lymph node metastases, and distant metastases. Wilcoxon signed-rank test was performed to compare the diagnostic accuracy of PET/CT and PET/MR. Result According to the standard of reference, a primary lesion was found in 14 patients. In 16 patients, the primary lesion remained occult. In the remaining 13 patients, lesions proved to be benign. PET/MR was superior to PET/CT for primary tumor detection (sensitivity/specificity, 0.85/0.97 vs 0.69/0.73; P = 0.020) and comparable to PET/CT for the detection of lymph node metastases (sensitivity/specificity, 0.93/1.00 vs 0.93/0.93; P = 0.157) and distant metastases (sensitivity/specificity, 1.00/0.97 vs 0.82/1.00; P = 0.564). PET/CT tended to misclassify physiologic FDG uptake as malignancy compared with PET/MR (8 patients vs 1 patient). Conclusions PET/MR outperforms PET/CT in the workup of suspected occult malignancies. PET/MR may replace PET/CT to improve clinical workflow.

Reduction of 18F-FDG Dose in Clinical PET/MR Imaging by Using Silicon Photomultiplier Detectors

Purpose To determine the level of clinically acceptable reduction in injected fluorine 18 (18F) fluorodeoxyglucose (FDG) dose in time-of-flight (TOF)-positron emission tomography(PET)/magnetic resonance (MR) imaging by using silicon photomultiplier (SiPM) detectors compared with TOF-PET/computed tomography (CT) using Lu1.8Y0.2SiO5(Ce), or LYSO, detectors in patients with different body mass indexes (BMIs). Materials and Methods Patients were enrolled in this study as part of a larger prospective study with a different purpose than evaluated in this study (NCT02316431). All patients gave written informed consent prior to inclusion into the study. In this study, 74 patients with different malignant diseases underwent sequential whole-body TOF-PET/CT and TOF-PET/MR imaging. PET images with simulated reduction of injected 18F-FDG doses were generated by unlisting the list-mode data from PET/MR imaging. Two readers rated the image quality of whole-body data sets, as well as the image quality in each body compartment, and evaluated the conspicuity of malignant lesions. Results The image quality with 70% or 60% of the injected dose of 18F-FDG at PET/MR imaging was comparable to that at PET/CT. With 50% of the injected dose, comparable image quality was maintained among patients with a BMI of less than 25 kg/m2. PET images without TOF reconstruction showed higher artifact scores and deteriorated sharpness than those with TOF reconstruction. Conclusion Sixty percent of the usually injected 18F-FDG dose (reduction of up to 40%) in patients with a BMI of more than 25 kg/m2 results in clinically adequate PET image quality in TOF-PET/MR imaging performed by using SiPM detectors. Additionally, in patients with a BMI of less than 25 kg/m2, 50% of the injected dose may safely be used.

Clinical evaluation of a block sequential regularized expectation maximization reconstruction algorithm in 18F-FDG PET/CT studies

Purpose To investigate the clinical performance of a block sequential regularized expectation maximization (BSREM) penalized likelihood reconstruction algorithm in oncologic PET/computed tomography (CT) studies. Methods A total of 410 reconstructions of 41 fluorine-18 fluorodeoxyglucose-PET/CT studies of 41 patients with a total of 2010 lesions were analyzed by two experienced nuclear medicine physicians. Images were reconstructed with BSREM (with four different &bgr; values) or ordered subset expectation maximization (OSEM) algorithm with/without time-of-flight (TOF/non-TOF) corrections. OSEM reconstruction postfiltering was 4.0 mm full-width at half-maximum; BSREM did not use postfiltering. Evaluation of general image quality was performed with a five-point scale using maximum intensity projections. Artifacts (category 1), image sharpness (category 2), noise (category 3), and lesion detectability (category 4) were analyzed using a four-point scale. Size and maximum standardized uptake value (SUVmax) of lesions were measured by a third reader not involved in the image evaluation. Results BSREM-TOF reconstructions showed the best results in all categories, independent of different body compartments. In all categories, BSREM non-TOF reconstructions were significantly better than OSEM non-TOF reconstructions (P<0.001). In almost all categories, BSREM non-TOF reconstruction was comparable to or better than the OSEM-TOF algorithm (P<0.001 for general image quality, image sharpness, noise, and P=1.0 for artifact). Only in lesion detectability was OSEM-TOF significantly better than BSREM non-TOF (P<0.001). Both BSREM-TOF and BSREM non-TOF showed a decreasing SUVmax with increasing &bgr; values (P<0.001) and TOF reconstructions showed a significantly higher SUVmax than non-TOF reconstructions (P<0.001). Conclusion The BSREM reconstruction algorithm showed a relevant improvement compared with OSEM reconstruction in PET/CT studies in all evaluated categories. BSREM might be used in clinical routine in conjunction with TOF to achieve better/higher image quality and lesion detectability or in PET/CT-systems without TOF-capability for enhancement of overall image quality as well.

Local resectability assessment of head and neck cancer: Positron emission tomography/MRI versus positron emission tomography/CT

The purpose of this study was to compare the diagnostic accuracy of positron emission tomography (PET)/MRI with PET/CT for local resectability of head and neck cancer.

Pitfalls and Limitations in Simultaneous PET/MRI

Simultaneous PET/MRI was introduced into the commercial market only a few years ago, and its availability is currently gaining momentum with the introduction of a second-generation PET/MRI system from an additional vendor. Furthermore, there is still an increasing interest in its potential in clinical and research applications. Despite very early technical infancy problems, which meanwhile have been solved, there are still different limitations that have to be worked around in daily routine responsibly by the physicists and physicians. This article gives an overview over the most common technical, logistical, and clinical limitations; artifacts; and pitfalls, without any claim for completeness. The readers will not only learn the background of the limitation but also partly learn about possible solutions. At the end of each paragraph, the readers will find a short summary for an easier overview of the topics discussed.

The effect of susceptibility artifacts related to metal implants on adjacent lesion assessment in simultaneous TOF PET/MR.

Metalic implants may affect attenuation correction (AC) in PET/MR imaging. The purpose of this study was to evaluate the effect of susceptibility artifacts related to metallic implants on adjacent metabolically active lesions in clinical simultaneous PET/MR scanning for both time-of-flight (TOF) and non-TOF reconstructed PET images. Methods: We included 27 patients without implants but with confirmed 18F-FDG–avid lesions adjacent to common implant locations. In all patients, a clinically indicated whole-body 18F-FDG PET/MR scan was acquired. Baseline non-TOF and TOF PET images were reconstructed. Reconstruction was repeated after the introduction of artificial signal voids in the AC map to simulate metallic implants in standard anatomic areas. All reconstructed images were qualitatively and quantitatively assessed and compared with the baseline images. Results: In total, 51 lesions were assessed. In 40 and 50 of these cases (non-TOF and TOF, respectively), the detectability of the lesions did not change; in 9 and 1 cases, the detectability changed; and in 2 non-TOF cases, the lesions were no longer visible after the introduction of metallic artifacts. The inclusion of TOF information significantly reduced artifacts due to simulated implants in the femoral head, sternum, and spine (P = 0.01, 0.01, and 0.03, respectively). It also improved image quality in these locations (P = 0.02, 0.01, and 0.01, respectively). The mean percentage error was −3.5% for TOF and −4.8% for non-TOF reconstructions, meaning that the inclusion of TOF information reduced the percentage error in SUVmax by 28.5% (P < 0.01). Conclusion: Qualitatively, there was a significant reduction of artifacts in the femoral head, sternum, and spine. There was also a significant qualitative improvement in image quality in these locations. Furthermore, our study indicated that simulated susceptibility artifacts related to metallic implants have a significant effect on small, moderately 18F-FDG–avid lesions near the implant site that possibly may go unnoticed without TOF information. On larger, highly 18F-FDG–avid lesions, the metallic implants had only a limited effect. The largest significant quantitative difference was found in artifacts of the sternum. There was only a weak inverse correlation between lesions affected by artifacts and distance from the implant.