Optimal Scaling Approaches for Perfusion MRI with Distorted Arterial Input Function (AIF) in Patients with Ischemic Stroke

Background: Diagnosis and timely treatment of ischemic stroke depends on the fast and accurate quantification of perfusion parameters. Arterial input function (AIF) describes contrast agent concentration over time as it enters the brain through the brain feeding artery. AIF is the central quantity required to estimate perfusion parameters. Inaccurate and distorted AIF, due to partial volume effects (PVE), would lead to inaccurate quantification of perfusion parameters. Methods: Fifteen patients suffering from stroke underwent perfusion MRI imaging at the Tri-Service General Hospital, Taipei. Various degrees of the PVE were induced on the AIF and subsequently corrected using rescaling methods. Results: Rescaled AIFs match the exact reference AIF curve either at peak height or at tail. Inaccurate estimation of CBF values estimated from non-rescaled AIFs increase with increasing PVE. Rescaling of the AIF using all three approaches resulted in reduced deviation of CBF values from the reference CBF values. In most cases, CBF map generated by rescaled AIF approaches show increased CBF and Tmax values on the slices in the left and right hemispheres. Conclusion: Rescaling AIF by VOF approach seems to be a robust and adaptable approach for correction of the PVE-affected multivoxel AIF. Utilizing an AIF scaling approach leads to more reasonable absolute perfusion parameter values, represented by the increased mean CBF/Tmax values and CBF/Tmax images.

Quantitative Myocardial Perfusion with 2D and 3D Sequences Alternating Every Heartbeat

PurposeTo evaluate a myocardial perfusion acquisition that alternates 2D simultaneous multi-slice (SMS) and 3D stack-of-stars (SoS) acquisitions each heartbeat. MethodsA hybrid saturation recovery radial 2D SMS and a saturation recovery 3D SoS sequence were created for the quantification of myocardial blood flow (MBF). Initial studies were done to study the effects of using only every other beat for the 2D SMS in two subjects, and for the 3D SoS in two subjects. Alternating heartbeat 2D SMS and 3D SoS were then performed in ten dog studies at rest, four dog studies at adenosine stress, and two human resting studies. 2D SMS acquisition acquired three slices and 3D SoS acquired six slices. An arterial input function (AIF) for 2D SMS was obtained using the first 24 rays. For 3D, the AIF was obtained in a 2D slice prior to each 3D SoS readout. Quantitative MBF analysis was performed for 2D SMS and 3D SoS separately, using a two-compartment model. ResultsAcquiring every-other-beat data resulted in 5-20% perfusion changes at rest for both 2D SMS and 3D SoS methods. For alternating acquisitions, 2D SMS and 3D SoS quantitative perfusion values were comparable for both the twelve rest studies (2D SMS: 0.68±0.15 vs 3D: 0.69±0.15 ml/g/min, p=0.85) and the four stress studies (2D SMS: 1.28±0.22 vs 3D: 1.30±0.24 ml/g/min, p=0.66).ConclusionEvery-other-beat acquisition changed estimated perfusion values relatively little for both sequences. 2D SMS and 3D SoS gave similar quantitative perfusion estimates when used in an alternating every-other-heartbeat acquisition. Such an approach allows consideration of more diverse perfusion acquisitions that could have complementary features, although testing in a cardiac disease population is needed.

Characterization in Non-human Primates of (R)-[18F]OF-Me-NB1 and (S)-[18F]OF-Me-NB1 for Imaging the GluN2B Subunits of the NMDA Receptor

PurposeGluN2B containing N-methyl-D-aspartate receptors (NMDARs) play an essential role in neurotransmission and are a potential treatment target for multiple neurological and neurodegenerative diseases, including stroke, Alzheimer’s disease, and Parkinson’s disease. In a previous report (R)-[18F]OF-Me-NB1 was reported to be more specific and selective for targeting the GluN2B subunits of the NMDAR. Here we report a comprehensive evaluation of (R)-[18F]OF-Me-NB1 and (S)-[18F]OF-Me-NB1 in non-human primates.MethodsThe radiosynthesis of (R)-[18F]OF-Me-NB1 and (S)-[18F]OF-Me-NB1 was accomplished as reported previously with minor modifications. PET scans in two rhesus monkeys were conducted on the Focus 220 scanner. Blocking studies were performed after treatment of the animals with the GluN2B antagonist Co101,244 or the sigma-1 receptor antagonist FTC-146. One-tissue (1T) compartment model and multilinear analysis-1 (MA1) method with arterial input function were used to obtain regional volume of distribution (VT, mL/cm3). Occupancy values by the two blockers were obtained by the Lassen plot. Regional non-displaceable binding potential (BPND) was calculated from the corresponding baseline VT and the VND derived from the occupancy plot of the Co101,244 blocking scans. Results(R)- and (S)-[18F]OF-Me-NB1 were produced in >99% radiochemical and enantiomeric purity, with molar activity of 224.22 ± 161.69 MBq/nmol at the end of synthesis (n=10). Metabolism was moderate, with ~30% parent compound remaining for (R)-[18F]OF-Me-NB1 and 20% for (S)-[18F]OF-Me-NB1 at 30 min post injection. Plasma free fraction was 1-2%. In brain regions both (R)- and (S)-[18F]OF-Me-NB1 displayed fast uptake with slower clearance for the (R)- than (S)-enantiomer. For (R)-[18F]OF-Me-NB1, both the 1T model and MA1 method gave reliable estimates of regional VT values, with MA1 VT (mL/cm3) ranging from 8.9 in the cerebellum to 12.8 in the cingulate cortex. Blocking with 0.25 mg/kg of Co101,244 greatly reduced the uptake of (R)-[18F]OF-Me-NB1 across all brain regions, resulting in an occupancy of 77% and VND of 6.36, while 0.027 mg/kg of FTC-146 reduced specific binding by 30%. Regional BPND, as a measure of specific binding signals, ranged from 0.40 in the cerebellum to 1.01 in the cingulate cortex.ConclusionsIn rhesus monkeys, (R)-[18F]OF-Me-NB1 exhibited fast kinetics and heterogeneous uptake across brain regions, while the (S)-enantiomer displayed a narrower dynamic range of uptake across regions. Blocking study with a GluN2B antagonist indicated binding specificity. Value of BPND were >0.5 in most brain regions, suggesting good in vivo specific binding signals. Taken together, results from the current study demonstrated the potential of (R)-[18F]OF-Me-NB1 as a useful radiotracer for imaging the GluN2B receptor.

Prediction of Tissue Damage Using a User-Independent Machine Learning Algorithm vs. Tmax Threshold Maps

(1) Background: To test the accuracy of a fully automated stroke tissue estimation algorithm (FASTER) to predict final lesion volumes in an independent dataset in patients with acute stroke; (2) Methods: Tissue-at-risk prediction was performed in 31 stroke patients presenting with a proximal middle cerebral artery occlusion. FDA-cleared perfusion software using the AHA recommendation for the Tmax threshold delay was tested against a prediction algorithm trained on an independent perfusion software using artificial intelligence (FASTER). Following our endovascular strategy to consequently achieve TICI 3 outcome, we compared patients with complete reperfusion (TICI 3) vs. no reperfusion (TICI 0) after mechanical thrombectomy. Final infarct volume was determined on a routine follow-up MRI or CT at 90 days after the stroke; (3) Results: Compared to the reference standard (infarct volume after 90 days), the decision forest algorithm overestimated the final infarct volume in patients without reperfusion. Underestimation was observed if patients were completely reperfused. In cases where the FDA-cleared segmentation was not interpretable due to improper definitions of the arterial input function, the decision forest provided reliable results; (4) Conclusions: The prediction accuracy of automated tissue estimation depends on (i) success of reperfusion, (ii) infarct size, and (iii) software-related factors introduced by the training sample. A principal advantage of machine learning algorithms is their improved robustness to artifacts in comparison to solely threshold-based model-dependent software. Validation on independent datasets remains a crucial condition for clinical implementations of decision support systems in stroke imaging.

Enhancement-constrained acceleration: A robust reconstruction framework in breast DCE-MRI

In patients with dense breasts or at high risk of breast cancer, dynamic contrast enhanced MRI (DCE-MRI) is a highly sensitive diagnostic tool. However, its specificity is highly variable and sometimes low; quantitative measurements of contrast uptake parameters may improve specificity and mitigate this issue. To improve diagnostic accuracy, data need to be captured at high spatial and temporal resolution. While many methods exist to accelerate MRI temporal resolution, not all are optimized to capture breast DCE-MRI dynamics. We propose a novel, flexible, and powerful framework for the reconstruction of highly-undersampled DCE-MRI data: enhancement-constrained acceleration (ECA). Enhancement-constrained acceleration uses an assumption of smooth enhancement at small time-scale to estimate points of smooth enhancement curves in small time intervals at each voxel. This method is tested in silico with physiologically realistic virtual phantoms, simulating state-of-the-art ultrafast acquisitions at 3.5s temporal resolution reconstructed at 0.25s temporal resolution (demo code available here). Virtual phantoms were developed from real patient data and parametrized in continuous time with arterial input function (AIF) models and lesion enhancement functions. Enhancement-constrained acceleration was compared to standard ultrafast reconstruction in estimating the bolus arrival time and initial slope of enhancement from reconstructed images. We found that the ECA method reconstructed images at 0.25s temporal resolution with no significant loss in image fidelity, a 4x reduction in the error of bolus arrival time estimation in lesions (p < 0.01) and 11x error reduction in blood vessels (p < 0.01). Our results suggest that ECA is a powerful and versatile tool for breast DCE-MRI.

A Short Dynamic Scan Method of Measuring Bone Metabolic Flux Using [18F]NaF PET

[18F]NaF PET measurements of bone metabolic flux (Ki) are conventionally obtained with 60-min dynamic scans analysed using the Hawkins model. However, long scan times make this method expensive and uncomfortable for subjects. Therefore, we evaluated and compared measurements of Ki with shorter scan times analysed with fixed values of the Hawkins model rate constants. The scans were acquired in a trial in 30 postmenopausal women, half treated with teriparatide (TPT) and half untreated. Sixty-minute PET-CT scans of both hips were acquired at baseline and week 12 after injection with 180 MBq [18F]NaF. Scans were analysed using the Hawkins model by fitting bone time–activity curves at seven volumes of interest (VOIs) with a semi-population arterial input function. The model was re-run with fixed rate-constants for dynamic scan times from 0–12 min increasing in 4-min steps up to 0–60 min. Using the Hawkins model with fixed rate-constants, Ki measurements with statistical power equivalent or superior to conventionally analysed 60-min dynamic scans were obtained with scan times as short as 12 min.

Variation in arterial input function in a large multicenter computed tomography perfusion study

Objectives To report the variation in computed tomography perfusion (CTP) arterial input function (AIF) in a multicenter stroke study and to assess the impact this has on CTP results. Methods CTP datasets from 14 different centers were included from the DUtch acute STroke (DUST) study. The AIF was taken as a direct measure to characterize contrast bolus injection. Statistical analysis was applied to evaluate differences in amplitude, area under the curve (AUC), bolus arrival time (BAT), and time to peak (TTP). To assess the clinical relevance of differences in AIF, CTP acquisitions were simulated with a realistic anthropomorphic digital phantom. Perfusion parameters were extracted by CTP analysis using commercial software (IntelliSpace Portal (ISP), version 10.1) as well as an in-house method based on block-circulant singular value decomposition (bSVD). Results A total of 1422 CTP datasets were included, ranging from 6 to 322 included patients per center. The measured values of the parameters used to characterize the AIF differed significantly with approximate interquartile ranges of 200–750 HU for the amplitude, 2500–10,000 HU·s for the AUC, 0–17 s for the BAT, and 10–26 s for the TTP. Mean infarct volumes of the phantom were significantly different between centers for both methods of perfusion analysis. Conclusions Although guidelines for the acquisition protocol are often provided for centers participating in a multicenter study, contrast medium injection protocols still vary. The resulting volumetric differences in infarct core and penumbra may impact clinical decision making in stroke diagnosis. Key Points • The contrast medium injection protocol may be different between stroke centers participating in a harmonized multicenter study. • The contrast medium injection protocol influences the results of X-ray computed tomography perfusion imaging. • The contrast medium injection protocol can impact stroke diagnosis and patient selection for treatment.