Single PET Scan Research Articles

Kinetic model-informed deep learning for multiplexed PET image separation

BackgroundMultiplexed positron emission tomography (mPET) imaging can measure physiological and pathological information from different tracers simultaneously in a single scan. Separation of the multiplexed PET signals within a single PET scan is challenging due to the fact that each tracer gives rise to indistinguishable 511 keV photon pairs, and thus no unique energy information for differentiating the source of each photon pair.MethodsRecently, many applications of deep learning for mPET image separation have been concentrated on pure data-driven methods, e.g., training a neural network to separate mPET images into single-tracer dynamic/static images. These methods use over-parameterized networks with only a very weak inductive prior. In this work, we improve the inductive prior of the deep network by incorporating a general kinetic model based on spectral analysis. The model is incorporated, along with deep networks, into an unrolled image-space version of an iterative fully 4D PET reconstruction algorithm.ResultsThe performance of the proposed method was evaluated on a simulated brain image dataset for dual-tracer [18\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^{18}$$\\end{document}F]FDG+[11\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^{11}$$\\end{document}C]MET PET image separation. The results demonstrate that the proposed method can achieve separation performance comparable to that obtained with single-tracer imaging. In addition, the proposed method outperformed the model-based separation methods (the conventional voxel-wise multi-tracer compartment modeling method (v-MTCM) and the image-space dual-tracer version of the fully 4D PET image reconstruction algorithm (IS-F4D)), as well as a pure data-driven separation [using a convolutional encoder-decoder (CED)], with fewer training examples.ConclusionsThis work proposes a kinetic model-informed unrolled deep learning method for mPET image separation. In simulation studies, the method proved able to outperform both the conventional v-MTCM method and a pure data-driven CED with less training data.

Assessment of cerebral drug occupancy in humans using a single PET-scan: A [11C]UCB-J PET study

PurposeHere, we evaluate a PET displacement model with a Single-step and Numerical solution in healthy individuals using the synaptic vesicle glycoprotein (SV2A) PET-tracer [11C]UCB-J and the anti-seizure medication levetiracetam (LEV). We aimed to (1) validate the displacement model by comparing the brain LEV-SV2A occupancy from a single PET scan with the occupancy derived from two PET scans and the Lassen plot and (2) determine the plasma LEV concentration-SV2A occupancy curve in healthy individuals.MethodsEleven healthy individuals (five females, mean age 35.5 [range: 25–47] years) underwent two 120-min [11C]UCB-J PET scans where an LEV dose (5–30 mg/kg) was administered intravenously halfway through the first PET scan to partially displace radioligand binding to SV2A. Five individuals were scanned twice on the same day; the remaining six were scanned once on two separate days, receiving two identical LEV doses. Arterial blood samples were acquired to determine the arterial input function and plasma LEV concentrations. Using the displacement model, the SV2A-LEV target engagement was calculated and compared with the Lassen plot method. The resulting data were fitted with a single-site binding model.ResultsSV2A occupancies and VND estimates derived from the displacement model were not significantly different from the Lassen plot (p = 0.55 and 0.13, respectively). The coefficient of variation was 14.6% vs. 17.3% for the Numerical and the Single-step solution in Bland-Altman comparisons with the Lassen plot. The average half maximal inhibitory concentration (IC50), as estimated from the area under the curve of the plasma LEV concentration, was 12.5 µg/mL (95% CI: 5–25) for the Single-Step solution, 11.8 µg/mL (95% CI: 4–25) for the Numerical solution, and 6.3 µg/mL (95% CI: 0.08-21) for the Lassen plot. Constraining Emax to 100% did not significantly improve model fits.ConclusionPlasma LEV concentration vs. SV2A occupancy can be determined in humans using a single PET scan displacement model. The average concentration of the three computed IC50 values ranges between 6.3 and 12.5 µg/mL. The next step is to use the displacement model to evaluate LEV occupancy and corresponding plasma concentrations in relation to treatment efficacy.Clinical trial registrationNCT05450822. Retrospectively registered 5 July 2022 https://clinicaltrials.gov/ct2/results? term=NCT05450822&Search=Search.

Deep learned triple-tracer multiplexed PET myocardial image separation.

In multiplexed positron emission tomography (mPET) imaging, physiological and pathological information from different radiotracers can be observed simultaneously in a single dynamic PET scan. The separation of mPET signals within a single PET scan is challenging due to the fact that the PET scanner measures the sum of the PET signals of all the tracers. The conventional multi-tracer compartment modeling (MTCM) method requires staggered injections and assumes that the arterial input functions (AIFs) of each tracer are known. In this work, we propose a deep learning-based method to separate triple-tracer PET images without explicitly knowing the AIFs. A dynamic triple-tracer noisy MLEM reconstruction was used as the network input, and dynamic single-tracer noisy MLEM reconstructions were used as training labels. A simulation study was performed to evaluate the performance of the proposed framework on triple-tracer ([ ]FDG+ +[ ]sestamibi) PET myocardial imaging. The results show that the proposed methodology substantially reduced the noise level compared to the results obtained from single-tracer imaging. Additionally, it achieved lower bias and standard deviation in the separated single-tracer images compared to the MTCM-based method at both the voxel and region of interest (ROI) levels. As compared to MTCM separation, the proposed method uses spatiotemporal information for separation, which improves the separation performance at both the voxel and ROI levels. The simulation study also demonstrates the feasibility and potential of the proposed DL-based method for the application to pre-clinical and clinical studies.

Novel voxelwise residual analysis of [11C]raclopride PET data improves detection of low-amplitude dopamine release

Existing methods for voxelwise transient dopamine (DA) release detection rely on explicit kinetic modeling of the [11C]raclopride PET time activity curve, which at the voxel level is typically confounded by noise, leading to poor performance for detection of low-amplitude DA release-induced signals. Here we present a novel data-driven, task-informed method—referred to as Residual Space Detection (RSD)—that transforms PET time activity curves to a residual space where DA release-induced perturbations can be isolated and processed. Using simulations, we demonstrate that this method significantly increases detection performance compared to existing kinetic model-based methods for low-magnitude DA release (simulated +100% peak increase in basal DA concentration). In addition, results from nine healthy controls injected with a single bolus of [11C]raclopride performing a finger tapping motor task are shown as proof-of-concept. The ability to detect relatively low magnitudes of dopamine release in the human brain using a single bolus injection, while achieving higher statistical power than previous methods, may additionally enable more complex analyses of neurotransmitter systems. Moreover, RSD is readily generalizable to multiple tasks performed during a single PET scan, further extending the capabilities of task-based single-bolus protocols.

Assessment of perfusion deficit with early phases of [18F]PI-2620 tau-PET versus [18F]flutemetamol-amyloid-PET recordings

PurposeCharacteristic features of amyloid-PET (A), tau-PET (T), and FDG-PET (N) can serve for the A/T/N classification of neurodegenerative diseases. Recent studies showed that the early, perfusion-weighted phases of amyloid- or tau-PET recordings serve to detect cerebrometabolic deficits equally to FDG-PET, therefore providing a surrogate of neuronal injury. As such, two channels of diagnostic information can be obtained in the setting of a single PET scan. However, there has hitherto been no comparison of early-phase amyloid- and tau-PET as surrogates for deficits in perfusion/metabolism. Therefore, we undertook to compare [18F]flutemetamol-amyloid-PET and [18F]PI-2620 tau-PET as “one-stop shop” dual purpose tracers for the detection of neurodegenerative disease.MethodsWe obtained early-phase PET recordings with [18F]PI-2620 (0.5–2.5 min p.i.) and [18F]flutemetamol (0–10 min p.i.) in 64 patients with suspected neurodegenerative disease. We contrasted global mean normalized images (SUVr) in the patients with a normal cohort of 15 volunteers without evidence of increased pathology to β-amyloid- and tau-PET examinations. Regional group differences of tracer uptake (z-scores) of 246 Brainnetome volumes of interest were calculated for both tracers, and the correlations of the z-scores were evaluated using Pearson’s correlation coefficient. Lobar compartments, regions with significant neuronal injury (z-scores < − 3), and patients with different neurodegenerative disease entities (e.g., Alzheimer’s disease or 4R-tauopathies) served for subgroup analysis. Additionally, we used partial regression to correlate regional perfusion alterations with clinical scores in cognition tests.ResultsThe z-scores of perfusion-weighted images of both tracers showed high correlations across the brain, especially in the frontal and parietal lobes, which were the brain regions with pronounced perfusion deficit in the patient group (R = 0.83 ± 0.08; range, 0.61–0.95). Z-scores of individual patients correlated well by region (R = 0.57 ± 0.15; range, 0.16–0.90), notably when significant perfusion deficits were present (R = 0.66 ± 0.15; range, 0.28–0.90).ConclusionThe early perfusion phases of [18F]PI-2620 tau- and [18F]flutemetamol-amyloid-PET are roughly equivalent indices of perfusion defect indicative of regional and lobar neuronal injury in patients with various neurodegenerative diseases. As such, either tracer may serve for two diagnostic channels by assessment of amyloid/tau status and neuronal activity.

Evaluation of in vivo staging of amyloid deposition in cognitively unimpaired elderly aged 78–94

Amyloid-beta (Aβ) deposition is common in cognitively unimpaired (CU) elderly >85 years. This study investigated amyloid distribution and evaluated three published in vivo amyloid-PET staging schemes from a cognitively unimpaired (CU) cohort aged 84.9 ± 4.3 years (n = 75). SUV-based principal component analysis (PCA) was applied to 18F-flutemetamol PET data to determine an unbiased regional covariance pattern of tracer uptake across grey matter regions. PET staging schemes were applied to the data and compared to the PCA output. Concentration of p-tau181 was measured in blood plasma. The PCA revealed three distinct components accounting for 91.2% of total SUV variance. PC1 driven by the large common variance of uptake in neocortical and striatal regions was significantly positively correlated with global SUVRs, APOE4 status and p-tau181 concentration. PC2 represented mainly non-specific uptake in typical amyloid-PET reference regions, and PC3 the occipital lobe. Application of the staging schemes demonstrated that the majority of the CU cohort (up to 93%) were classified as having pathological amount and distribution of Aβ. Good correspondence existed between binary (+/−) classification and later amyloid stages, however, substantial differences existed between schemes for low stages with 8–17% of individuals being unstageable, i.e., not following the sequential progression of Aβ deposition. In spite of the difference in staging outcomes there was broad spatial overlap between earlier stages and PC1, most prominently in default mode network regions. This study critically evaluated the utility of in vivo amyloid staging from a single PET scan in CU elderly and found that early amyloid stages could not be consistently classified. The majority of the cohort had pathological Aβ, thus, it remains an open topic what constitutes abnormal brain Aβ in the oldest-old and what is the best method to determine that.

Measurement of Striatal Dopamine Release Induced by Neuropsychological Stimulation in Positron Emission Tomography With Dual Injections of [11C]Raclopride.

ObjectivesPositron emission tomography (PET) with [11C]raclopride has been applied to measure changes in the concentration of endogenous dopamine induced by pharmacological challenge or neuropsychological stimulation by evaluating the binding potential (BP) between the baseline and activated state. Recently, to reliably estimate BP in the activated state, a new approach with dual-bolus injections in a single PET scan was developed. In this study, we investigated the feasibility of applying this dual-bolus injection approach to measure changes in endogenous dopamine levels induced by cognitive tasks in humans.MethodsFirst, the reproducibility of BP estimation using the dual-bolus injection approach was evaluated using PET scans without stimulation in nine healthy volunteers. A 90-min scan was performed with bolus injections of [11C]raclopride administered at the beginning of the scan and 45 min after the first injection. BPs in the striatum for the first injection (BP1) and second injection (BP2) were estimated using an extended simplified reference tissue model, and the mean absolute difference (MAD) between the two BPs was calculated. The MAD was also compared with the conventional bolus-plus-continuous infusion approach. Next, PET studies with a cognitive reinforcement learning task were performed on 10 healthy volunteers using the dual-bolus injection approach. The BP1 at baseline and BP2 at the activated state were estimated, and the reduction in BP was evaluated.ResultsIn the PET scans without stimulation, the dual-bolus injection approach showed a smaller MAD (<2%) between BP1 and BP2 than the bolus-plus-continuous infusion approach, demonstrating good reproducibility of this approach. In the PET scans with the cognitive task performance, the reduction in BP was not observed in the striatum by either approach, showing that the changes in dopamine level induced by the cognitive tasks performed in this study were not sufficient to be detected by PET.ConclusionOur results indicate that the cognitive task-induced changes in dopamine-related systems may be complex and difficult to measure accurately using PET scans. However, the proposed dual-bolus injection approach provided reliable BP estimates with high reproducibility, suggesting that it has the potential to improve the accuracy of PET scans for measuring changes in dopamine concentrations.

Extra-striatal dopamine in Parkinson's disease with rapid eye movement sleep behavior disorder.

Rapid eye movement sleep behavior disorder (RBD) is a common condition found in more than 50% of the patients with Parkinson's disease (PD). Molecular imaging shows that PD with RBD (PD-RBD+) have lower striatal dopamine transporter activity within the caudate and putamen relative to PD without RBD (PD-RBD-). However, the characterization of the extra-striatal dopamine within the mesocortical and mesolimbic pathways remains unknown. We aim to elucidate this with PET imaging in 15 PD-RBD+ and 15 PD-RBD- patients, while having 15 age-matched healthy controls (HC). Each participant underwent a single PET scan with [11 C]FLB-457 to detect the D2 receptor availability within the extra-striatal regions of interest (ROI), including the prefrontal, temporal, and limbic areas. [11 C]FLB-457 retention was expressed as the nondisplaceable binding potential. Our results reveal that relative to HC, PD-RBD+ and PD-RBD- patients have lower levels of D2 receptor availability within the uncus parahippocampus, superior, lateral, and inferior temporal cortex. PD-RBD+ showed steep decline in D2 receptors within the left uncus parahippocampus with increasing disease severity, but this was not observed for PD-RBD- patients. Findings imply that extra-striatal dopaminergic system may play a role in contributing to symptomatic progress in PD patients with RBD. However, validation with more advanced PD patients are needed.

Trajectory of clinical symptoms accelerates after 10 years of amyloid

AbstractBackgroundUnderstanding the time between amyloid biomarker detection and the onset of cognitive and functional symptoms will be critical in making care planning decisions and determining an optimal time period for potential interventions. Previous studies have suggested that amyloid, as measured with Pittsburg compound B (PiB), can be detected 15‐20 years before symptom onset. However, the designs of these studies do not—on an individual basis—provide an understanding of the duration between amyloid accumulation and symptom onset. Recent research from several groups has suggested that the rate of amyloid accumulation is consistent across individuals, making it possible to estimate the chronicity of amyloid positivity from a single PET scan. In relation to this measure of amyloid chronicity, we examined trajectories of cognitive and functional symptoms, as measured by the Clinical Dementia Rating Scale (CDR).MethodThe sample included 347 study visits from 84 amyloid positive (A+) participants enrolled in the Wisconsin Registry for Alzheimer’s Prevention or the Wisconsin Alzheimer’s Disease Research Center (Table 1). For participant visits that administered the Quick Dementia Rating System (QDRS), the equivalent CDR sum of boxes (CDR‐SB) score was calculated. A mixed effects model with random slope and intercept was used to evaluate the association between chronicity and CDR‐SB (with covariates APOE4 carriage, sex and baseline age). To investigate potential non‐linearity in chronicity’s association with CDR‐SB, a chronicity‐by‐baseline chronicity category (&lt;0, 0‐5, 5‐10, 10‐15, 15‐20 and &gt;20 years) interaction was examined (&lt;0=reference range).ResultSpaghetti plots display the trajectories of CDR‐SB values in relation to amyloid chronicity; shading indicates clinical guidelines for interpreting CDR‐SB (Figure 1). The results of the model indicate that the trajectory of CDR‐SB is significantly greater after 10 years of amyloid chronicity (Table 2 &amp; Figure 2).ConclusionLargely consistent with previous studies, the results indicate that clinical symptoms begin to emerge after 10 years of amyloid positivity. However, symptom severity appeared to vary greatly between 10 – 30 years of amyloid chronicity. Future research will aim to identify other processes that may contribute to decline as well as genetic and modifiable factors that may promote resiliency and delay the onset of clinical symptoms.

Quantification of myocardial blood flow with 11C-hydroxyephedrine dynamic PET: comparison with 15O-H2O PET

Background11C-hydroxyephedrine (HED) PET has been used to evaluate the myocardial sympathetic nervous system (SNS). Here we sought to establish a simultaneous approach for quantifying both myocardial blood flow (MBF) and the SNS from a single HED PET scan. MethodsTen controls and 13 patients with suspected cardiac disease were enrolled. The inflow rate of 11C-HED (K1) was obtained using a one-tissue-compartment model. We compared this rate with the MBF derived from 15O-H2O PET. In the controls, the relationship between K1 from 11C-HED PET and the MBF from 15O-H2O PET was linked by the Renkin-Crone model. ResultsThe relationship between K1 from 11C-HED PET and the MBF from 15O-H2O PET from the controls’ data was approximated as follows: K1 = (1 − 0.891 * exp(− 0.146/MBF)) * MBF. In the validation set, the correlation coefficient demonstrated a significantly high relationship for both the whole left ventricle (r = 0.95, P < 0.001) and three coronary territories (left anterior descending artery: r = 0.96, left circumflex artery: r = 0.81, right coronary artery: r = 0.86; P < 0.001, respectively). Conclusion11C-HED can simultaneously estimate MBF and sympathetic nervous function without requiring an additional MBF scan for assessing mismatch areas between MBF and SNS.

Bayesian Estimation of the ntPET Model in Single-Scan Competition PET Studies.

This paper proposes an innovative method, named b-ntPET, for solving a competition model in PET. The model is built upon the state-of-the-art method called lp-ntPET. It consists in identifying the parameters of the PET kinetic model relative to a reference region that rule the steady state exchanges, together with the identification of four additional parameters defining a displacement curve caused by an endogenous neurotransmitter discharge, or by a competing injected drug targeting the same receptors as the PET tracer. The resolution process of lp-ntPET is however suboptimal due to the use of discretized basis functions, and is very sensitive to noise, limiting its sensitivity and accuracy. Contrary to the original method, our proposed resolution approach first estimates the probability distribution of the unknown parameters using Markov-Chain Monte-Carlo sampling, distributions from which the estimates are then inferred. In addition, and for increased robustness, the noise level is jointly estimated with the parameters of the model. Finally, the resolution is formulated in a Bayesian framework, allowing the introduction of prior knowledge on the parameters to guide the estimation process toward realistic solutions. The performance of our method was first assessed and compared head-to-head with the reference method lp-ntPET using well-controlled realistic simulated data. The results showed that the b-ntPET method is substantially more robust to noise and much more sensitive and accurate than lp-ntPET. We then applied the model to experimental animal data acquired in pharmacological challenge studies and human data with endogenous releases induced by transcranial direct current stimulation. In the drug challenge experiment on cats using [18F]MPPF, a serotoninergic 1A antagonist radioligand, b-ntPET measured a dose response associated with the amount of the challenged injected concurrent 5-HT1A agonist, where lp-ntPET failed. In human [11C]raclopride experiment, contrary to lp-ntPET, b-ntPET successfully detected significant endogenous dopamine releases induced by the stimulation. In conclusion, our results showed that the proposed method b-ntPET has similar performance to lp-ntPET for detecting displacements, but with higher resistance to noise and better robustness to various experimental contexts. These improvements lead to the possibility of detecting and characterizing dynamic drug occupancy from a single PET scan more efficiently.

A Monte Carlo approach for improving transient dopamine release detection sensitivity

Current methods using a single PET scan to detect voxel-level transient dopamine release—using F-test (significance) and cluster size thresholding—have limited detection sensitivity for clusters of release small in size and/or having low release levels. Specifically, simulations show that voxels with release near the peripheries of such clusters are often rejected—becoming false negatives and ultimately distorting the F-distribution of rejected voxels. We suggest a Monte Carlo method that incorporates these two observations into a cost function, allowing erroneously rejected voxels to be accepted under specified criteria. In simulations, the proposed method improves detection sensitivity by up to 50% while preserving the cluster size threshold, or up to 180% when optimizing for sensitivity. A further parametric-based voxelwise thresholding is then suggested to better estimate the release dynamics in detected clusters. We apply the Monte Carlo method to a pilot scan from a human gambling study, where additional parametrically unique clusters are detected as compared to the current best methods—results consistent with our simulations.

Automatic calculation of myocardial external efficiency using a single 11C-acetate PET scan

BackgroundMyocardial external efficiency (MEE) is defined as the ratio of kinetic energy associated with cardiac work [forward cardiac output (FCO)*mean systemic pressure] and the chemical energy from oxygen consumed (MVO2) by the left ventricular mass (LVM). We developed a fully automated method for estimating MEE based on a single 11C-acetate PET scan without ECG-gating. Methods and ResultsTen healthy controls, 34 patients with aortic valve stenosis (AVS), and 20 patients with mitral valve regurgitation (MVR) were recruited in a dual-center study. MVO2 was calculated using washout of 11C -acetate activity. FCO and LVM were calculated automatically using dynamic PET and parametric image formation. FCO and LVM were also obtained using cardiac magnetic resonance (CMR) in all subjects. The correlation between MEEPET-CMR and MEEPET was high (r = 0.85, P < 0.001) without significant bias. MEEPET was 23.6 ± 4.2% for controls and was lowered in AVS (17.2 ± 4.3%, P < 0.001) and in MVR (18.0 ± 5.2%, P = 0.004). MEEPET was strongly associated with both NYHA class (P < 0.001) and the magnitude of valvular dysfunction (mean aortic gradient: P < 0.001, regurgitant fraction: P = 0.009). ConclusionA single 11C-acetate PET yields accurate and automated MEE results on different scanners. MEE might provide an unbiased measurement of the phenotypic response to valvular disease.

The diagnostic accuracy of detecting malignant transformation of low-grade glioma using O-(2-[18F]fluoroethyl)-l-tyrosine positron emission tomography: a retrospective study.

The diagnostic accuracy of O-(2-[18F]fluoroethyl)-l-tyrosine (FET) PET scanning in detecting the malignant transformation of low-grade gliomas (LGGs) is controversial. In this study, the authors retrospectively assessed the diagnostic potential of FET PET in patients with MRI-suspected malignant progression of LGGs that had previously been treated and the relationship between FET uptake and MRI and molecular biomarkers. Forty-two patients who had previously undergone surgical or multimodal treatment for a histologically verified LGG were referred for FET PET assessment because of clinical signs and/or MRI findings suggestive of tumor progression. Maximal and mean tumor-to-brain ratios (TBRmax and TBRmean, respectively) on FET PET as well as kinetic FET PET parameters (time to peak [TTP] and time-activity curve [TAC]) were determined. Final diagnoses were confirmed histologically. The diagnostic accuracy of FET parameters, separately and combined, for the detection of malignant progression was evaluated using receiver operating characteristic (ROC) curve analysis. Possible predictors that might influence the diagnostic accuracy of FET PET were assessed using multiple linear regression analysis. Spearman’s rank correlation r method was applied to determine the correlation between TBRmax and TAC, and molecular biomarkers from tumor tissues. A total of 47 FET PET scans were obtained and showed no significant association between FET parameters and contrast enhancement on MRI. ROC curve analyses overall were unable to demonstrate any significant differentiation between nontransformed LGGs and LGGs that had transformed to high-grade gliomas when evaluating FET parameters separately or combined. After excluding the oligodendroglial subgroup, a significant difference was observed between nontransformed and transformed LGGs when combining FET parameters (i.e., TBRmax > 1.6, TAC describing a plateau or decreasing pattern, and TTP < 25 minutes), with the best result yielded by a combined analysis of TBRmax > 1.6 and TAC with a plateau or decreasing pattern (sensitivity 75% and specificity 83%, p = 0.003). The difference was even greater when patients who had previously undergone oncological treatment were also excluded (sensitivity 93% and specificity 100%, p = 0.001). Multiple linear regression analysis revealed that the presence of an oligodendroglial component (p = 0.029), previous oncological treatment (p = 0.039), and the combined FET parameters (p = 0.027) were significant confounding factors in the detection of malignant progression. TBRmax was positively correlated with increasing cell density (p = 0.040) and inversely correlated with IDH1 mutation (p = 0.006). A single FET PET scan obtained at the time of radiological and/or clinical progression seems to be of limited value in distinguishing transformed from nontransformed LGGs, especially if knowledge of the primary tumor histopathology is not known. Therefore, FET PET imaging alone is not adequate to replace histological confirmation, but it may provide valuable information on the location and delineation of active tumor tissue, as well as an assessment of tumor biology in a subgroup of LGGs.

Florbetapir-PET β-amyloid imaging and associated neuropsychological trajectories in survivors of critical illness: A case series

PurposeCognitive impairment resembling Alzheimer's disease is common in survivors of critical illness. We hypothesized that Intensive Care Unit (ICU) survivors with cognitive impairment would have significant amyloid and designed a pilot study to explore this relationship. Materials and methodsA pilot, case series of a convenience sample of 14 adult medical and surgical ICU survivors, in a clinical neuroradiology clinic. Patients underwent cognitive testing at 3months, 1year, 4years, and 6years after hospital discharge with the Repeatable Battery for the Assessment of Neuropsychological Status. They received a single PET scan using amyloid PET imaging (florbetapir F18) 2 to 4years after their ICU stay. ResultsAmyloid (defined as a Standard Uptake Value ratio or SUVr >1.10) was present in 2 of 14 (14%) individuals, both of whom demonstrated significant cognitive impairment yet no consistent decline over time. Of the 6 impaired patients (RBANS<78), 4 (66.7%) were amyloid negative. ConclusionsIt is feasible to assess ICU survivors with amyloid imaging. In this small sample, most patients with cognitive impairment were negative on amyloid PET imaging, which raises the possibility that ICU survivors may experience a unique form of dementia not driven by an amyloid related mechanism.

Rapid computation of single PET scan rest-stress myocardial blood flow parametric images by table look up.

We have recently reported a method for measuring rest-stress myocardial blood flow (MBF) using a single, relatively short, PET scan session. The method requires two IV tracer injections, one to initiate rest imaging and one at peak stress. We previously validated absolute flow quantitation in ml/min/cc for standard bull's eye, segmental analysis. In this work, we extend the method for fast computation of rest-stress MBF parametric images. We provide an analytic solution to the single-scan rest-stress flow model which is then solved using a two-dimensional table lookup method (LM). Simulations were performed to compare the accuracy and precision of the lookup method with the original nonlinear method (NLM). Then the method was applied to 16 single scan rest/stress measurements made in 12 pigs: seven studied after infarction of the left anterior descending artery (LAD) territory, and nine imaged in the native state. Parametric maps of rest and stress MBF as well as maps of left (fLV ) and right (fRV ) ventricular spill-over fractions were generated. Regions of interest (ROIs) for 17 myocardial segments were defined in bull's eye fashion on the parametric maps. The mean of each ROI was then compared to the rest (K1r ) and stress (K1s ) MBF estimates obtained from fitting the 17 regional TACs with the NLM. In simulation, the LM performed as well as the NLM in terms of precision and accuracy. The simulation did not show that bias was introduced by the use of a predefined two-dimensional lookup table. In experimental data, parametric maps demonstrated good statistical quality and the LM was computationally much more efficient than the original NLM. Very good agreement was obtained between the mean MBF calculated on the parametric maps for each of the 17 ROIs and the regional MBF values estimated by the NLM (K1mapLM =1.019×K1ROINLM +0.019, R2 =0.986; mean difference=0.034±0.036mL/min/cc). We developed a table lookup method for fast computation of parametric imaging of rest and stress MBF. Our results show the feasibility of obtaining good quality MBF maps using modest computational resources, thus demonstrating that the method can be applied in a clinical environment to obtain full quantitative MBF information.

Rapid processing of PET list-mode data for efficient uncertainty estimation and data analysis

In this technical note we propose a rapid and scalable software solution for the processing of PET list-mode data, which allows the efficient integration of list mode data processing into the workflow of image reconstruction and analysis. All processing is performed on the graphics processing unit (GPU), making use of streamed and concurrent kernel execution together with data transfers between disk and CPU memory as well as CPU and GPU memory. This approach leads to fast generation of multiple bootstrap realisations, and when combined with fast image reconstruction and analysis, it enables assessment of uncertainties of any image statistic and of any component of the image generation process (e.g. random correction, image processing) within reasonable time frames (e.g. within five minutes per realisation). This is of particular value when handling complex chains of image generation and processing.The software outputs the following: (1) estimate of expected random event data for noise reduction; (2) dynamic prompt and random sinograms of span-1 and span-11 and (3) variance estimates based on multiple bootstrap realisations of (1) and (2) assuming reasonable count levels for acceptable accuracy. In addition, the software produces statistics and visualisations for immediate quality control and crude motion detection, such as: (1) count rate curves; (2) centre of mass plots of the radiodistribution for motion detection; (3) video of dynamic projection views for fast visual list-mode skimming and inspection; (4) full normalisation factor sinograms. To demonstrate the software, we present an example of the above processing for fast uncertainty estimation of regional SUVR (standard uptake value ratio) calculation for a single PET scan of 18F-florbetapir using the Siemens Biograph mMR scanner.

Comparison of Early-Phase 11C-Deuterium-l-Deprenyl and 11C-Pittsburgh Compound B PET for Assessing Brain Perfusion in Alzheimer Disease.

The PET tracer (11)C-deuterium-L-deprenyl ((11)C-DED) has been used to visualize activated astrocytes in vivo in patients with Alzheimer disease (AD). In this multitracer PET study, early-phase (11)C-DED and (11)C-Pittsburgh compound B ((11)C-PiB) (eDED and ePiB, respectively) were compared as surrogate markers of brain perfusion, and the extent to which (11)C-DED binding is influenced by brain perfusion was investigated. (11)C-DED, (11)C-PiB, and (18)F-FDG dynamic PET scans were obtained in age-matched groups comprising AD patients (n = 8), patients with mild cognitive impairment (n = 17), and healthy controls (n = 16). A modified reference Patlak model was used to quantify (11)C-DED binding. A simplified reference tissue model was applied to both (11)C-DED and (11)C-PiB to measure brain perfusion relative to the cerebellar gray matter (R1) and binding potentials. (11)C-PiB retention and (18)F-FDG uptake were also quantified as target-to-pons SUV ratios in 12 regions of interest (ROIs). The strongest within-subject correlations with the corresponding R1 values (R1,DED and R1,PiB, respectively) and with (18)F-FDG uptake were obtained when the eDED and ePiB PET data were measured 1-4 min after injection. The optimum eDED/ePiB intervals also showed strong, significant ROI-based intersubject Pearson correlations with R1,DED/R1,PiB and with (18)F-FDG uptake, whereas (11)C-DED binding was largely independent of brain perfusion, as measured by eDED. Corresponding voxelwise correlations confirmed the ROI-based results. Temporoparietal eDED or ePiB brain perfusion measurements were highly discriminative between patient and control groups, with discriminative ability statistically comparable to that of temporoparietal (18)F-FDG glucose metabolism. Hypometabolism extended over wider regions than hypoperfusion in patient groups compared with controls. The 1- to 4-min early-frame intervals of (11)C-DED or (11)C-PiB are suitable surrogate measures for brain perfusion. (11)C-DED binding is independent of brain perfusion, and thus (11)C-DED PET can provide information on both functional (brain perfusion) and pathologic (astrocytosis) aspects from a single PET scan. In comparison with glucose metabolism, early-phase (11)C-DED and (11)C-PiB perfusion appear to provide complementary rather than redundant information.

Predicting Response to Neoadjuvant Chemotherapy with PET Imaging Using Convolutional Neural Networks.

Imaging of cancer with 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) has become a standard component of diagnosis and staging in oncology, and is becoming more important as a quantitative monitor of individual response to therapy. In this article we investigate the challenging problem of predicting a patient’s response to neoadjuvant chemotherapy from a single 18F-FDG PET scan taken prior to treatment. We take a “radiomics” approach whereby a large amount of quantitative features is automatically extracted from pretherapy PET images in order to build a comprehensive quantification of the tumor phenotype. While the dominant methodology relies on hand-crafted texture features, we explore the potential of automatically learning low- to high-level features directly from PET scans. We report on a study that compares the performance of two competing radiomics strategies: an approach based on state-of-the-art statistical classifiers using over 100 quantitative imaging descriptors, including texture features as well as standardized uptake values, and a convolutional neural network, 3S-CNN, trained directly from PET scans by taking sets of adjacent intra-tumor slices. Our experimental results, based on a sample of 107 patients with esophageal cancer, provide initial evidence that convolutional neural networks have the potential to extract PET imaging representations that are highly predictive of response to therapy. On this dataset, 3S-CNN achieves an average 80.7% sensitivity and 81.6% specificity in predicting non-responders, and outperforms other competing predictive models.

Simultaneous functional imaging using fPET and fMRI.

Brain mapping of task-associated changes in metabolism with PET has been accomplished by subtracting scans acquired during two distinct static states. We have demonstrated that PET can provide truly dynamic information on cerebral energy metabolism using constant infusion of FDG and multiple stimuli in a single experiment. We demonstrate here that the functional PET (fPET-FDG) method accomplished simultaneously with fMRI, can enable the first direct comparisons in time, space and magnitude of hemodynamics and oxygen and glucose consumption. The imaging studies were performed on a 3T Tim-Trio MR scanner modified to support an MR-compatible BrainPET insert. Ten healthy subjects were included. The total PET acquisition and infusion time was 90 minutes. We did 3 blocks of right hand fingers tapping for 10 minutes at 30, 50 and 70 minutes after the beginning of the PET acquisition. ASL and BOLD imaging were acquired simultaneously during the motor paradigm. Changes in glucose utilization are easily observed as changes in the TAC slope of the PET data (FDG utilization rate) and in the derivative signal during motor stimuli in the activated voxels. PET and MRI (ASL, and BOLD) activations are largely colocalized but with very different statistical significance and temporal dynamic, especially in the ipsilateral side of the stimuli. This study demonstrated that motor activation can be measured dynamically during a single FDG PET scan. The complementary nature of fPET-FDG to fMRI capitalizes on the emerging technology of hybrid MR-PET scanners. fPET-FDG, combined with quantitative fMRI methods, allow us to simultaneously measure dynamic changes in glucose utilization and hemodynamic, addressing vital questions about neurovascular coupling.