Evaluation of 3-dimensional stereotactic surface projection rendering of arterial spin labeling data in a clinical cohort.

BACKGROUND AND PURPOSE: To assess the feasibility of 3-dimensional stereotactic surface projection (3D-SSP) as applied to arterial spin labeling (ASL) in a clinical pilot study. METHODS: A retrospective sample of 10 consecutive patients who underwent ASL as part of a clinically indicated MR examination was collected during this pilot study. Five additional subjects with normal cerebral perfusion served as a control group. Following voxel-wise M0-correction, cerebral blood flow (CBF) quantification, and stereotactic anatomic standardization, voxel-wise CBF from an individual's ASL dataset was extracted to a set of predefined surface pixels (3D-SSP). A normal database was created from averaging the extracted CBF datasets of the control group. Patients' datasets were compared individually with the normal database by calculating a Z-score on a pixel-by-pixel basis and were displayed in 3D-SSP views for visual inspection. Independent, two-expert reader assessment, using a 3-point scale, compared standard quantitative CBF images to the 3D-SSP maps. RESULTS: Patterns and severities of regionally reduced CBF were identified, by both independent readers, in the 3D-SSP maps. Reader assessment demonstrated preference for 3D-SSP over traditionally displayed standard quantitative CBF images in three of four evaluated imaging metrics (p = .026, .031, and .013, respectively); 3D-SSP maps were never found to be inferior to the standard quantitative CBF images. CONCLUSIONS: Three-dimensional SSP maps are feasible in a clinical population and enable quantitative data extraction and localization of perfusion abnormalities by means of stereotactic coordinates in a condensed display. The proposed method is a promising approach for interpreting cerebrovascular pathophysiology.

The RSNA QIBA Profile for Amyloid PET as an Imaging Biomarker for Cerebral Amyloid Quantification.

A standardized approach to acquiring amyloid PET images increases their value as disease and drug response biomarkers. Most 18F PET amyloid brain scans often are assessed only visually (per regulatory labels), with a binary decision indicating the presence or absence of Alzheimer disease amyloid pathology. Minimizing technical variance allows precise, quantitative SUV ratios (SUVRs) for early detection of ß-amyloid plaques and allows the effectiveness of antiamyloid treatments to be assessed with serial studies. Methods: The Quantitative Imaging Biomarkers Alliance amyloid PET biomarker committee developed and validated a profile to characterize and reduce the variability of SUVRs, increasing statistical power for these assessments. Results: On achieving conformance, sites can justify a claim that brain amyloid burden reflected by the SUVR is measurable to a within-subject coefficient of variation of no more than 1.94% when the same radiopharmaceutical, scanner, acquisition, and analysis protocols are used. Conclusion: This overview explains the claim, requirements, barriers, and potential future developments of the profile to achieve precision in clinical and research amyloid PET imaging.

Deep-learning prediction of amyloid deposition from early-phase amyloid positron emission tomography imaging.

OBJECTIVE: While the use of biomarkers for the detection of early and preclinical Alzheimer's Disease has become essential, the need to wait for over an hour after injection to obtain sufficient image quality can be challenging for patients with suspected dementia and their caregivers. This study aimed to develop an image-based deep-learning technique to generate delayed uptake patterns of amyloid positron emission tomography (PET) images using only early-phase images obtained from 0-20 min after radiotracer injection. METHODS: We prepared pairs of early and delayed [11C]PiB dynamic images from 253 patients (cognitively normal n = 32, fronto-temporal dementia n = 39, mild cognitive impairment n = 19, Alzheimer's disease n = 163) as a training dataset. The neural network was trained with the early images as the input, and the output was the corresponding delayed image. A U-net convolutional neural network (CNN) and a conditional generative adversarial network (C-GAN) were used for the deep-learning architecture and the data augmentation methods, respectively. Then, an independent test data set consisting of early-phase amyloid PET images (n = 19) was used to generate corresponding delayed images using the trained network. Two nuclear medicine physicians interpreted the actual delayed images and predicted delayed images for amyloid positivity. In addition, the concordance of the actual delayed and predicted delayed images was assessed statistically. RESULTS: The concordance of amyloid positivity between the actual versus AI-predicted delayed images was 79%(κ = 0.60) and 79% (κ = 0.59) for each physician, respectively. In addition, the physicians' agreement rate was at 89% (κ = 0.79) when the same image was interpreted. And, the actual versus AI-predicted delayed images were not readily distinguishable (correct answer rate, 55% and 47% for each physician, respectively). The statistical comparison of the actual versus the predicted delated images indicated that the peak signal-to-noise ratio (PSNR) was 21.8 dB ± 2.2 dB, and the structural similarity index (SSIM) was 0.45 ± 0.04. CONCLUSION: This study demonstrates the feasibility of an image-based deep-learning framework to predict delayed patterns of Amyloid PET uptake using only the early phase images. This AI-based image generation method has the potential to reduce scan time for amyloid PET and increase the patient throughput, without sacrificing diagnostic accuracy for amyloid positivity.

18F-FDG PET Imaging in Neurodegenerative Dementing Disorders: Insights into Subtype Classification, Emerging Disease Categories, and Mixed Dementia with Copathologies.

Since the invention of 18F-FDG as a neurochemical tracer in the 1970s, 18F-FDG PET has been used extensively for dementia research and clinical applications. FDG, a glucose analog, is transported into the brain via glucose transporters and metabolized in a concerted process involving astrocytes and neurons. Although the exact cellular mechanisms of glucose consumption are still under investigation, 18F-FDG PET can sensitively detect altered neuronal activity due to neurodegeneration. Various neurodegenerative disorders affect different areas of the brain, which can be depicted as altered 18F-FDG uptake by PET. The spatial patterns and severity of such changes can be reproducibly visualized by statistical mapping technology, which has become widely available in the clinic. The differentiation of 3 major neurodegenerative disorders by 18F-FDG PET, Alzheimer disease (AD), frontotemporal dementia (FTD), and dementia with Lewy bodies (DLB), has become standard practice. As the nosology of FTD evolves, frontotemporal lobar degeneration, the umbrella term for pathology affecting the frontal and temporal lobes, has been subclassified clinically into behavioral variant FTD; primary progressive aphasia with 3 subtypes, semantic, nonfluent, and logopenic variants; and movement disorders including progressive supranuclear palsy and corticobasal degeneration. Each of these subtypes is associated with differential 18F-FDG PET findings. The discovery of new pathologic markers and clinicopathologic correlations via larger autopsy series have led to newly recognized or redefined disease categories, such as limbic-predominant age-related TDP-43 encephalopathy, hippocampus sclerosis, primary age-related tauopathy, and argyrophilic grain disease, which have become a focus of investigations by molecular imaging. These findings need to be integrated into the modern interpretation of 18F-FDG PET. Recent pathologic investigations also have revealed a high prevalence, particularly in the elderly, of mixed dementia with overlapping and coexisting pathologies. The interpretation of 18F-FDG PET is evolving from a traditional dichotomous diagnosis of AD versus FTD (or DLB) to a determination of the most predominant underlying pathology that would best explain the patient's symptoms, for the purpose of care guidance. 18F-FDG PET is a relatively low cost and widely available imaging modality that can help assess various neurodegenerative disorders in a single test and remains the workhorse in clinical dementia evaluation.

An Unusual Presentation of Breast Implant Rupture.

ABSTRACT: A 99mTc-methylene diphosphate (MDP) bone scintigraphy scan was performed to follow up metastatic breast cancer. A circular region of radiotracer uptake in the location of the patient's left breast implant was seen. Review of the same-day chest CT and the prior CT showed that the patient's breast implant had ruptured in the interim. The 99mTc-MDP uptake in the capsule of the breast implant was attributed to rupture, likely secondary to inflammation. This is an example of an unusual presentation of breast implant rupture discovered on a 99mTc-MDP bone scintigraphy scan.

Application of artificial intelligence in brain molecular imaging.

Initial development of artificial Intelligence (AI) and machine learning (ML) dates back to the mid-twentieth century. A growing awareness of the potential for AI, as well as increases in computational resources, research, and investment are rapidly advancing AI applications to medical imaging and, specifically, brain molecular imaging. AI/ML can improve imaging operations and decision making, and potentially perform tasks that are not readily possible by physicians, such as predicting disease prognosis, and identifying latent relationships from multi-modal clinical information. The number of applications of image-based AI algorithms, such as convolutional neural network (CNN), is increasing rapidly. The applications for brain molecular imaging (MI) include image denoising, PET and PET/MRI attenuation correction, image segmentation and lesion detection, parametric image formation, and the detection/diagnosis of Alzheimer's disease and other brain disorders. When effectively used, AI will likely improve the quality of patient care, instead of replacing radiologists. A regulatory framework is being developed to facilitate AI adaptation for medical imaging.

Artificial Intelligence for Brain Molecular Imaging.

AI has been applied to brain molecular imaging for over 30 years. The past two decades, have seen explosive progress. AI applications span from operations processes such as attenuation correction and image generation, to disease diagnosis and prediction. As sophistication in AI software platforms increases, and the availability of large imaging data repositories become common, future studies will incorporate more multidimensional datasets and information that may truly reach "superhuman" levels in the field of brain imaging. However, even with a growing level of complexity, these advanced networks will still require human supervision for appropriate application and interpretation in medical practice.

Utility of SPECT Functional Neuroimaging of Pain.

Functional neuroimaging modalities vary in spatial and temporal resolution. One major limitation of most functional neuroimaging modalities is that only neural activation taking place inside the scanner can be imaged. This limitation makes functional neuroimaging in many clinical scenarios extremely difficult or impossible. The most commonly used radiopharmaceutical in Single Photon Emission Tomography (SPECT) functional brain imaging is Technetium 99 m-labeled Ethyl Cysteinate Dimer (ECD). ECD is a lipophilic compound with unique pharmacodynamics. It crosses the blood brain barrier and has high first pass extraction by the neurons proportional to regional brain perfusion at the time of injection. It reaches peak activity in the brain 1 min after injection and is then slowly cleared from the brain following a biexponential mode. This allows for a practical imaging window of 1 or 2 h after injection. In other words, it freezes a snapshot of brain perfusion at the time of injection that is kept and can be imaged later. This unique feature allows for designing functional brain imaging studies that do not require the patient to be inside the scanner at the time of brain activation. Functional brain imaging during severe burn wound care is an example that has been extensively studied using this technique. Not only does SPECT allow for imaging of brain activity under extreme pain conditions in clinical settings, but it also allows for imaging of brain activity modulation in response to analgesic maneuvers whether pharmacologic or non-traditional such as using virtual reality analgesia. Together with its utility in extreme situations, SPECTS is also helpful in investigating brain activation under typical pain conditions such as experimental controlled pain and chronic pain syndromes.

Intranasal Paclitaxel Alters Alzheimer's Disease Phenotypic Features in 3xTg-AD Mice.

BACKGROUND: Microtubule stabilizing drugs, commonly used as anti-cancer therapeutics, have been proposed for treatment of Alzheimer's disease (AD); however, many do not cross the blood-brain barrier. OBJECTIVE: This research investigated if paclitaxel (PTX) delivered via the intranasal (IN) route could alter the phenotypic progression of AD in 3xTg-AD mice. METHODS: We administered intranasal PTX in 3XTg-AD mice (3xTg-AD n = 15, 10 weeks and n = 10, 44 weeks, PTX: 0.6 mg/kg or 0.9%saline (SAL)) at 2-week intervals. After treatment, 3XTg-AD mice underwent manganese-enhanced magnetic resonance imaging to measure in vivo axonal transport. In a separate 3XTg-AD cohort, PTX-treated mice were tested in a radial water tread maze at 52 weeks of age after four treatments, and at 72 weeks of age, anxiety was assessed by an elevated-plus maze after 14 total treatments. RESULTS: PTX increased axonal transport rates in treated 3XTg-AD compared to controls (p≤0.003). Further investigation using an in vitro neuron model of Aß-induced axonal transport disruption confirmed PTX prevented axonal transport deficits. Confocal microscopy after treatment found fewer phospho-tau containing neurons (5.25±3.8 versus 8.33±2.5, p < 0.04) in the CA1, altered microglia, and reduced reactive astrocytes. PTX improved performance of 3xTg-AD on the water tread maze compared to controls and not significantly different from WT (Day 5, 143.8±43 versus 91.5±77s and Day 12, 138.3±52 versus 107.7±75s for SAL versus PTX). Elevated plus maze revealed that PTX-treated 3xTg-AD mice spent more time exploring open arms (Open arm 129.1±80 versus 20.9±31s for PTX versus SAL, p≤0.05). CONCLUSION: Taken collectively, these findings indicate that intranasal-administered microtubule-stabilizing drugs may offer a potential therapeutic option for treating AD.

Technical Note: Quantification of blood-spinal cord barrier permeability after application of magnetic resonance-guided focused ultrasound in spinal cord injury.

PURPOSE: To demonstrate that magnetic resonance-guided focused ultrasound (MRgFUS) facilitates blood-spinal cord barrier (BSCB) permeability and develop observer-independent MRI quantification of BSCB permeability after MRgFUS for spinal cord injury (SCI). METHODS: Noninjured Sprague-Dawley rats (n = 3) underwent MRgFUS and were administered Evans blue post-MRgFUS to confirm BSCB opening. Absorbance was measured by spectrophotometry and correlated with its corresponding image intensity. Rats (n = 21) underwent T8-T10 laminectomy and extradural compression of the spinal cord (23g weighted aneurysm-type clip, 1 min). The intervention group (n = 11) was placed on a preclinical MRgFUS system, administered microbubbles (Optison, 0.2 mL/kg), and received 3 MRgFUS sonications (25 ms bursts, 1 Hz pulses for 3 min, 3 acoustic W, approximately 1.0-2.1 MPa peak pressure as measured via hydrophone). The sham group (n = 10) received equivalent procedures with no sonications. T1w MRI was obtained both pre- and post-MRgFUS BSCB opening. Spinal cords were segmented manually or semiautomatically and a Pearson correlation with P ≤ 0.001 was used to correlate the two segmentation methods. MRgFUS sonication and control regions intensity values were evaluated with a paired t-test with a P ≤ 0.01. RESULTS: Semiautomatic segmentation reduced computational time by 95% and was correlated with manual segmentation (Pearson = 0.92, P < 0.001, n = 71 regions). In the noninjured rat group, Evans blue absorbance correlated with image intensity in the MRgFUS and control regions (Pearson = 0.82, P = 0.02, n = 6). In rats that underwent the SCI procedure, an increase in signal intensity in the MRgFUS targeted region relative to control was seen in all SCI rats (10.65 ± 12.4%, range: 0.96-43.9%, n = 11, P = 0.002). SCI sham MRgFUS revealed no change (0.63 ± 0.52%, 95% CI 0.320.95, n = 10). This result was significant between both groups (P = 0.003). CONCLUSION: The implemented semiautomatic segmentation procedure improved data analysis efficiency. Quantitative methods using contrast-enhanced MRI with histological validation are sensitive for detection of blood-spinal cord barrier opening induced by magnetic resonance-guided focused ultrasound.

Toward a Universal Readout for 18F-Labeled Amyloid Tracers: The CAPTAINs Study.

To date, 3 18F-labeled PET tracers have been approved for assessing cerebral amyloid plaque pathology in the diagnostic workup of suspected Alzheimer disease (AD). Although scanning protocols are relatively similar across tracers, U.S. Food and Drug Administration- and the European Medicines Agency-approved visual rating protocols differ among the 3 tracers. This proof-of-concept study assessed the comparability of the 3 approved visual rating protocols to classify a scan as amyloid-positive or -negative, when applied by groups of experts and nonexperts to all 3 amyloid tracers. Methods: In an international multicenter approach, both expert (n = 4) and nonexpert raters (n = 3) rated scans acquired with 18F-florbetaben, 18F-florbetapir and 18F-flutemetamol. Scans obtained with each tracer were presented for reading according to all 3 approved visual rating protocols. In a randomized order, every single scan was rated by each reader according to all 3 protocols. Raters were blinded for the amyloid tracer used and asked to rate each scan as positive or negative, giving a confidence judgment after each response. Percentage of visual reader agreement, interrater reliability, and agreement of each visual read with binary quantitative measures (fixed SUV ratio threshold for positive or negative scans) were computed. These metrics were analyzed separately for expert and nonexpert groups. Results: No significant differences in using the different approved visual rating protocols were observed across the different metrics of agreement in the group of experts. Nominal differences suggested that the 18F-florbetaben visual rating protocol achieved the highest interrater reliability and accuracy especially under low confidence conditions. For the group of nonexpert raters, significant differences between the different visual rating protocols were observed with overall moderate-to-fair accuracy and with the highest reliability for the 18F-florbetapir visual rating protocol. Conclusion: We observed high interrater agreement despite applying different visual rating protocols for all 18F-labeled amyloid tracers. This implies that the results of the visual interpretation of amyloid imaging can be well standardized and do not depend on the rating protocol in experts. Consequently, the creation of a universal visual assessment protocol for all amyloid imaging tracers appears feasible, which could benefit especially the less-experienced readers.

Brain [F-18]FDG PET for Clinical Dementia Workup: Differential Diagnosis of Alzheimer's Disease and Other Types of Dementing Disorders.

PET imaging with [F-18]FDG has been used extensively for research and clinical applications in dementia. In the brain, [F-18]FDG accumulates around synapses and represents local neuronal activity. Patterns of altered [F-18]FDG uptake reflecting local neuronal dysfunction provide differential diagnostic clues for various dementing disorders. Image interpretation can be accomplished by employing statistical brain mapping techniques. Various guidelines have been published to support the appropriate use of [F-18]FDG PET for clinical dementia workup. PET images with [F-18]FDG demonstrate distinct patterns of decreased uptake for Alzheimer's disease (AD), Dementia with Lewy bodies (DLB), and frontotemporal dementia (FTD) as well as its multiple subtypes such as behavioral variant FTD, primary progressive aphasia (PPA), progressive supranuclear palsy, and corticobasal degeneration to aid in the differential diagnoses. Mixed dementia, not only AD + Vascular Dementia, but also AD + other neurodegenerative disorders, should also be considered when interpreting [F-18]FDG PET images. Brain PET imaging with [F-18]FDG remains a valuable component of dementia workup owing to its relatively low cost, differential diagnostic performance, widespread availability, and physicians' experience over more than 40 years since the initial development.

Amyloid-PET and 18F-FDG-PET in the diagnostic investigation of Alzheimer's disease and other dementias.

Various biomarkers are available to support the diagnosis of neurodegenerative diseases in clinical and research settings. Among the molecular imaging biomarkers, amyloid-PET, which assesses brain amyloid deposition, and 18F-fluorodeoxyglucose (18F-FDG) PET, which assesses glucose metabolism, provide valuable and complementary information. However, uncertainty remains regarding the optimal timepoint, combination, and an order in which these PET biomarkers should be used in diagnostic evaluations because conclusive evidence is missing. Following an expert panel discussion, we reached an agreement on the specific use of the individual biomarkers, based on available evidence and clinical expertise. We propose a diagnostic algorithm with optimal timepoints for these PET biomarkers, also taking into account evidence from other biomarkers, for early and differential diagnosis of neurodegenerative diseases that can lead to dementia. We propose three main diagnostic pathways with distinct biomarker sequences, in which amyloid-PET and 18F-FDG-PET are placed at different positions in the order of diagnostic evaluations, depending on clinical presentation. We hope that this algorithm can support diagnostic decision making in specialist clinical settings with access to these biomarkers and might stimulate further research towards optimal diagnostic strategies.

Temporal muscle uptake causing an unusual focal artifact on three-dimensional stereotactic surface projections statistical maps of (18F) fluorodeoxyglucose brain positron emission tomography in a patient with Alzheimer's disease.

(18F) fluorodeoxyglucose brain positron emission tomography and statistical mapping analysis, such as three-dimensional stereotactic surface projections, have been used widely for the evaluation of dementia patients. We present an unusual focal artifact on the statistical maps resulting from intense temporal muscle uptake in a patient with Alzheimer's disease. Various degrees of physiologic uptake can be seen in head and neck muscles. However, it is unusual to see a focal artifact on the statistical maps due to temporal muscle uptake. This case illustrates the importance of quality control of imaging processing when atypical findings are seen on statistical maps.

The Role of the Cerebellum in Pain Perception: A Brain SPECT Perfusion Study in Patients with Burn Injuries.

BACKGROUND AND PURPOSE: Virtual reality (VR) is a promising tool for distraction analgesia. This study aims to compare brain perfusion patterns while patients were undergoing burn wound care in two conditions-VR distraction and control (NoVR). METHODS: With IRB approval, four patients hospitalized for acute burn care (three males and one female) participated in the study. All patients underwent wound care on two consecutive days; 1 day with standard analgesia and adjunctive VR, and the other day with standard analgesia alone, otherwise the wound care was very similar. Tc-99m ethyl cysteinate dimer was injected during wound care at the time of peak pain. Subjective patient reports on a 0-10 scale of pain intensity, time spent thinking about pain, and "fun" as well as opioid equivalent usage were analyzed. Voxel by voxel subtraction analysis of brain perfusion Single Photon Emission Computed Tomography (SPECT) images was performed at the group level. Statistical significance threshold was defined as P < .05. RESULTS: Mean group subjective scores (VR, NoVR, statistical significance, and P-value) were observed for maximal pain intensity (9.0, 8.8, insignificant, and P = .809), time spent thinking about pain (5.2, 10.0, significant, and P = .015), and fun (6.0, 2.5, significant, and P = .012). Subtraction group analysis demonstrated VR-induced modulation of brain activity with statistically significant relative suppression of cerebellar activation in the VR compared to intense cerebellar activation in the NoVR environments. CONCLUSION: Relative decrease in cerebellar perfusion based on stringent statistical threshold in the VR environment combined with improved subjective pain experience supports the hypotheses on the role of cerebellum in perception of noxious stimuli.