Tomography, Volume 4, Issue 3 (September 2018) – 7 articles

Multicenter clinical trials that use positron emission tomography (PET) imaging frequently rely on stable bias in imaging biomarkers to assess drug effectiveness. Many well-documented factors cause variability in PET intensity values. Two of the largest scanner-dependent errors are scanner calibration and reconstructed image resolution variations. For clinical trials, an increase in measurement error significantly increases the number of patient scans needed. We aim to provide a robust quality assurance system using portable PET/computed tomography “pocket” phantoms and automated image analysis algorithms with the goal of reducing PET measurement variability. A set of the “pocket” phantoms was scanned with patients, affixed to the underside of a patient bed. Our software analyzed the obtained images and estimated the image parameters. The analysis consisted of 2 steps, automated phantom detection and estimation of PET image resolution and global bias. Performance of the algorithm was tested under variations in image bias, resolution, noise, and errors in the expected sphere size. A web-based application was implemented to deploy the image analysis pipeline in a cloud-based infrastructure to support multicenter data acquisition, under Software-as-a-Service (SaaS) model. The automated detection algorithm localized the phantom reliably. Simulation results showed stable behavior when image properties and input parameters were varied. The PET “pocket” phantom has the potential to reduce and/or check for standardized uptake value measurement errors. Full article

This study evaluated the feasibility of using only diagnostically relevant magnetic resonance (MR) images together with deep learning for positron emission tomography (PET)/MR attenuation correction (deepMRAC) in the pelvis. Such an approach could eliminate dedicated MRAC sequences that have limited diagnostic utility but can substantially lengthen acquisition times for multibed position scans. We used axial T2 and T1 LAVA Flex magnetic resonance imaging images that were acquired for diagnostic purposes as inputs to a 3D deep convolutional neural network. The network was trained to produce a discretized (air, water, fat, and bone) substitute computed tomography (CT) (CT_sub). Discretized (CT_ref-discrete) and continuously valued (CT_ref) reference CT images were created to serve as ground truth for network training and attenuation correction, respectively. Training was performed with data from 12 subjects. CT_sub, CT_ref, and the system MRAC were used for PET/MR attenuation correction, and quantitative PET values of the resulting images were compared in 6 test subjects. Overall, the network produced CT_sub with Dice coefficients of 0.79 ± 0.03 for cortical bone, 0.98 ± 0.01 for soft tissue (fat: 0.94 ± 0.0; water: 0.88 ± 0.02), and 0.49 ± 0.17 for bowel gas when compared with CT_ref-discrete. The root mean square error of the whole PET image was 4.9% by using deepMRAC and 11.6% by using the system MRAC. In evaluating 16 soft tissue lesions, the distribution of errors for maximum standardized uptake value was significantly narrower using deepMRAC (−1.0% ± 1.3%) than using system MRAC method (0.0% ± 6.4%) according to the Brown–Forsy the test (P < .05). These results indicate that improved PET/MR attenuation correction can be achieved in the pelvis using only diagnostically relevant MR images. Full article

Amine chemical exchange saturation transfer (CEST) echoplanar imaging (EPI) provides unique pH and amino acid MRI contrast, enabling sensitive detection of altered microenvironment properties in various diseases. However, CEST contrast is sensitive to static magnetic field (B0) inhomogeneities. Here we propose 2 new B₀ correction algorithms for use in correcting pH-weighted amine CEST EPI based on k-means clustering and Lorentzian fitting of CEST data: the iterative downsampling estimation using Lorentzian fitting and the 2-stage Lorentzian estimation with 4D polynomial fitting. Higher quality images of asymmetric magnetization transfer ratio (MTR_asym) at 3.0 ppm could be obtained with the proposed algorithms than with the existing B₀ correction methods. In particular, the proposed methods are shown to improve the intertissue consistency, interpatient consistency, and tumor region signal-to-noise ratio of MTRasym at 3.0 ppm images, with nonexcessive computation time. Full article

Here, we developed a symmetric echo-planar spectroscopic imaging (EPSI) sequence for hyperpolarized ¹³C imaging on a clinical hybrid positron emission tomography/magnetic resonance imaging system. The pulse sequence uses parallel reconstruction pipelines to separately reconstruct data from odd-and-even gradient echoes to reduce artifacts from gradient imbalances. The ramp-sampled data in the spatiotemporal frequency space are regridded to compensate for the chemical-shift displacements. Unaliasing of nonoverlapping peaks outside of the sampled spectral width was performed to double the effective spectral width. The sequence was compared with conventional phase-encoded chemical-shift imaging (CSI) in phantoms, and it was evaluated in a canine cancer patient with ameloblastoma after injection of hyperpolarized [1-¹³C]pyruvate. The relative signal-to-noise ratio of EPSI with respect to CSI was 0.88, which is consistent with the decrease in sampling efficiency due to ramp sampling. Data regridding in the spatiotemporal frequency space significantly reduced spatial blurring compared with direct fast Fourier transform. EPSI captured the spatial distributions of both metabolites and their temporal dynamics in vivo with an in-plane spatial resolution of 5 × 9 mm² and a temporal resolution of 3 seconds. Significantly higher spatial and temporal resolution for delineating anatomical structures in vivo was achieved for EPSI metabolic maps than for CSI maps, which suffered spatiotemporal blurring. The EPSI sequence showed promising results in terms of short acquisition time and sufficient spectral bandwidth of 500 Hz, allowing to adjust the trade-off between signal-to-noise ratio and encoding speed. Full article

Numerous patient groups receive >1 medication and as such represent a potential point of improvement in today's healthcare setup, as the combined or cumulative effects are difficult to monitor in an individual patient. Here we show the ability to monitor the pharmacological effect of 2 classes of medications sequentially, namely, 2,4-dinitrophenol, a mitochondrial uncoupler, and dichloroacetate, a pyruvate dehydrogenase kinase inhibitor, both targeting the oxygen-dependent energy metabolism. We show that although the 2 drugs target 2 different metabolic pathways connected ultimately to oxygen metabolism, we could distinguish the 2 in vivo by using hyperpolarized [1-¹³C]pyruvate magnetic resonance imaging. A statistically significantly different pyruvate dehydrogenase flux was observed by reversing the treatment order of 2,4-dinitrophenol and dichloroacetate. The significance of this study is the demonstration of the ability to monitor the metabolic cumulative effects of 2 distinct therapeutics on an in vivo organ level using hyperpolarized magnetic resonance imaging. Full article

The recent report of the existence of meningeal lymphatic vessels (MLVs) in human and nonhuman primates used both histology and magnetic resonance imaging (MRI). Many questions about the physiology and function of these lymphatic vessels remain unanswered. Through the combination of appropriately positioned saturation bands and time-of-flight angiography sequences, MRI can resolve direction of flow within vessels without the use of exogenous contrast agent. Six healthy volunteers underwent high-resolution MRI of the MLVs running alongside the superior sagittal sinus to determine the direction of the lymphatic flow. In all subjects, the lymphatic flow was posterior to anterior, countercurrent to the direction of venous flow in the superior sagittal sinus and alongside the superior sagittal sinus. This flow strongly supports that a large proportion of the CNS lymphatic flow in humans is directed to the cribriform plate. The countercurrent direction of flow in the MLVs relative to venous flow in the superior sagittal sinus has implications for modeling flow of fluid and solutes across the various compartments of the CNS. A hypothetical compartmental model incorporating countercurrent flow is presented here. Full article

Magnetic resonance imaging (MRI) is an extremely useful tool for the detection and characterization of numerous pathologic processes. Although patients can benefit from the use of MRI, claustrophobia is a major issue in some cases. This fear alone can lead to cancellation of the scanning procedure in some cases and the need for conscious sedation in others. Patient anxiety during the scan can also lead to increased motion-related artifacts on the images with associated degradation of the diagnostic quality of the study. To alleviate these problems, our team developed a virtual reality (VR) tool (app) to educate patients about MRI and simulate the experience of actually being scanned. The app is totally immersive and incorporates both the visual and auditory sensations that patients encounter during an MRI scan. Patients also learn about potential conditions and implanted devices that may preclude the safe performance of the examination. This VR tool not only educates patients about MRI and its importance in their care, but also allows them to virtually experience what it is like to have a MRI scan. This technology has the potential to decrease both claustrophobic cancellations and patient anxiety before a MRI scan. Full article