Search

To develop artificial intelligence (AI) system that assists in checking endotracheal tube (ETT) placement on chest X-rays (CXRs) and evaluate whether it can move into clinical validation as a quality improvement tool.A retrospective data set including 2000 de-identified images from intensive care unit patients was split into 1488 for training and 512 for testing. AI was developed to automatically identify the ETT, trachea, and carina using semantically embedded neural networks that combine a declarative knowledge base with deep neural networks. To check the ETT tip placement, a "safe zone" was computed as the region inside the trachea and 3-7 cm above the carina. Two AI outputs were evaluated: (1) ETT overlay, (2) ETT misplacement alert messages. Clinically relevant performance metrics were compared against prespecified thresholds of85% overlay accuracy and positive predictive value (PPV)30% and negative predictive value NPV95% for alerts to move into clinical validation.An ETT was present in 285 of 512 test cases. The AI detected 95% (271/285) of ETTs, 233 (86%) of these with accurate tip localization. The system (correctly) did not generate an ETT overlay in 221/227 CXRs where the tube was absent for an overall overlay accuracy of 89% (454/512). The alert messages indicating that either the ETT was misplaced or not detected had a PPV of 83% (265/320) and NPV of 98% (188/192).The chest X-ray AI met prespecified performance thresholds to move into clinical validation.

Background 2-(4'- [11C]Methylaminophenyl)-6-hydroxybenzothiazole ([11C]-PiB), purportedly a specific imaging agent for cerebral amyloid-β plaques, is a specific, high affinity substrate for estrogen sulfotransferase (SULT1E1), an enzyme that regulates estrogen homeostasis. Objective In this work, we use positron emission tomography (PET) imaging with [11C]-PiB to assess the functional activity of SULT1E1 in the brain of moyamoya disease patients. Methods Ten moyamoya subjects and five control patients were evaluated with [11C]-PiB PET and structural MRI scans. Additionally, a patient with relapsing-remitting multiple sclerosis (RRMS) received [11C]-PiB PET scans before and after steroidal and immunomodulatory therapy. Parametric PET images were established to assess SULT1E1 distribution in the inflamed brain tissue. Results Increased [11C]-PiB SRTM DVR in the thalamus, pons, corona radiata, and internal capsule of moyamoya cohort subjects was observed in comparison with controls (p ≤ 0.01). This was observed in patients without treatment, with collateralization, and also after radiation. The post-treatment [11C]-PiB PET scan in one RRMS patient also revealed substantially reduced subcortical brain inflammation. In validation studies, [11C]-PiB autoradiography signal in the peri-infarct area of the rat middle cerebral arterial occlusion stroke model was shown to correlate with SULT1E1 immunohistochemistry. Conclusion Strong [11C]-PiB PET signal associated with intracranial inflammation in the moyamoya syndrome cohort and a single RRMS patient appears consistent with functional imaging of SULT1E1 activity in the human brain. This preliminary work offers substantial and direct evidence that significant [11C]-PiB PET focal signals can be obtained from the living human brain with intracranial inflammation, signals not attributable to amyloid-β plaques.

e21108 Background: RECIST 1.1 can underestimate treatment benefits of immunotherapy, with irRECIST and iRECIST accounting for atypical responses. Inter-reader discordances are known to occur in a dual reader paradigm. Our objective is to compare inter-reader reliability between RECIST 1.1, irRECIST, and iRECIST. Methods: This is a retrospective analysis of advanced NSCLC patients treated with pembrolizumab at our institution as part of the KEYNOTE-001 study. All trial imaging was interpreted by two radiologists. RECIST 1.1, irRECIST, and iRECIST categorical responses and agreement for progressive disease (PD) was compared by kappa statistic. Time to progression (TTP) or time to censor was compared between readers by paired t test. Relationship to disease progression and overall survival (OS) was assessed by log rank. Results: Of 98 patients, 77 had baseline and subsequent imaging available for 5.8 mean timepoints with 42.9 weeks of follow up. From this group, 45 patients had imaging beyond iUPD for confirmation and were analyzed. PD occurred by reader 1, reader 2 in 34, 33 patients by RECIST 1.1 (k = 0.591, CI = 0.320-0.863), 31, 29 patients by irRECIST (k = 0.501, CI = 0.234-0.768), and 27, 22 patients by iRECIST iCPD (confirmed-PD) (k = 0.690, CI = 0.485-0.896). There was no significant difference in reader agreement by RECIST 1.1, irRECIST, iRECIST (p = 0.38, 0.60, 0.26). There was a significant difference in time to progression between RECIST 1.1, irRECIST and iRECIST, with median PFS 3.4 months (2.6-4.6), 4.7 (3.5-6.8) and 8.7 (6.9-14.5) (p < 0.0001). PD by any criteria was not significantly correlated with OS. Conclusions: PD confirmation by iRECIST resulted in substantial reader agreement compared to moderate reader agreement by RECIST 1.1 and irRECIST. There were significant differences in TTP between the criteria, with iRECIST having the longest TTP. PD by each criteria did not correlate with a significant difference in OS.[Table: see text]

Caution is warranted when in vitro results of biomarkers labeled with tritium were perfunctorily used to criticize in vivo data and conclusions derived with the same tracers labeled with positron emitters and positron emission tomography (PET). This concept is illustrated herein with the PET utilization of [18F]FDDNP, a biomarker used for in vivo visualization of β-amyloid and tau protein neuroaggregates in humans, later contradicted by in vitro data reported with [3H]FDDNP. In this investigation, we analyze the multiple factors involved in the experimental design of the [3H]FDDNP in vitro study that led to the erroneous interpretation of results. The present work describes full details on the synthesis, characterization, purity, and kinetics of radiolytic stability of [3H]FDDNP. The optimal in vitro conditions for detecting tau and β-amyloid protein aggregates using macroscopic and microscopic autoradiography with both [18F]FDDNP and [3H]FDDNP are also presented. Macroscopic autoradiography determinations were performed with [3H]FDDNP of verified purity using established methods described previously in the literature. The autoradiographic results using phosphate buffered saline (PBS) with less than 1 % EtOH and pure, freshly prepared [3H]FDDNP compared with the earlier reported data using [3H]FDDNP of undetermined purity and PBS in 10 % EtOH demonstrate the critical importance of rigorous experimental design for meaningful in vitro determinations. [18F]FDDNP binding to both amyloid plaques and neurofibrillary tangles was confirmed by amyloid and tau immunohistochemical stains of adjacent tissues. This work illustrates the sensitivity of in vitro techniques to various experimental conditions and underscores that conclusions obtained from translational in vitro to in vivo determinations must always be performed with extreme care to avoid wrong interpretations that can be perpetuated and assumed without further analysis.

Author(s): Chen, Stephen T; Siddarth, Prabha; Merrill, David A; Martinez, Jacqueline; Emerson, Natacha D; Liu, Jie; Wong, Koon-Pong; Satyamurthy, Nagichettiar; Giza, Christopher C; Huang, Sung-Cheng; Fitzsimmons, Robert P; Bailes, Julian; Omalu, Bennet; Barrio, Jorge R; Small, Gary W | Abstract: BackgroundOur group has shown that in vivo tau brain binding patterns from FDDNP-PET scans in retired professional football players with suspected chronic traumatic encephalopathy differ from those of tau and amyloid aggregate binding observed in Alzheimer's disease (AD) patients and cognitively-intact controls.ObjectiveTo compare these findings with those from military personnel with histories of mild traumatic brain injury(mTBI).MethodsFDDNP-PET brain scans were compared among 7 military personnel and 15 retired players with mTBI histories and cognitive and/or mood symptoms, 24 AD patients, and 28 cognitively-intact controls. Nonparametric ANCOVAs with Tukey-Kramer adjusted post-hoc comparisons were used to test for significant differences in regional FDDNP binding among subject groups.ResultsFDDNP brain binding was higher in military personnel compared to controls in the amygdala, midbrain, thalamus, pons, frontal and anterior and posterior cingulate regions (p l 0.01-0.0001). Binding patterns in the military personnel were similar to those of the players except for the amygdala and striatum (binding higher in players; p = 0.02-0.003). Compared with the AD group, the military personnel showed higher binding in the midbrain (p = 0.0008) and pons (p = 0.002) and lower binding in the medial temporal, lateral temporal, and parietal regions (all p = 0.02).ConclusionThis first study of in vivo tau and amyloid brain signals in military personnel with histories of mTBI shows binding patterns similar to those of retired football players and distinct from the binding patterns in AD and normal aging, suggesting the potential value of FDDNP-PET for early detection and treatment monitoring in varied at-risk populations.

Caution is warranted when in vitro results of biomarkers labeled with tritium were perfunctorily used to criticize in vivo data and conclusions derived with the same tracers labeled with positron emitters and positron emission tomography (PET). This concept is illustrated herein with the PET utilization of [The present work describes full details on the synthesis, characterization, purity, and kinetics of radiolytic stability of [The autoradiographic results using phosphate buffered saline (PBS) with less than 1 % EtOH and pure, freshly prepared [This work illustrates the sensitivity of in vitro techniques to various experimental conditions and underscores that conclusions obtained from translational in vitro to in vivo determinations must always be performed with extreme care to avoid wrong interpretations that can be perpetuated and assumed without further analysis.

© 2017 The Authors Objective: Because curcumin's anti-inflammatory properties may protect the brain from neurodegeneration, we studied its effect on memory in non-demented adults and explored its impact on brain amyloid and tau accumulation using 2-(1-{6-[(2-[F-18]fluoroethyl)(methyl)amino]-2-naphthyl}ethylidene)malononitrile positron emission tomography (FDDNP-PET). Methods: Forty subjects (age 51–84 years) were randomized to a bioavailable form of curcumin (Theracurmin® containing 90 mg of curcumin twice daily [N = 21]) or placebo (N = 19) for 18 months. Primary outcomes were verbal (Buschke Selective Reminding Test [SRT]) and visual (Brief Visual Memory Test-Revised [BVMT-R]) memory, and attention (Trail Making A) was a secondary outcome. FDDNP-PET signals (15 curcumin, 15 placebo) were determined in amygdala, hypothalamus, medial and lateral temporal, posterior cingulate, parietal, frontal, and motor (reference) regions. Mixed effects general linear models controlling for age and education, and effect sizes (ES; Cohen's d) were estimated. Results: SRT Consistent Long-Term Retrieval improved with curcumin (ES = 0.63, p = 0.002) but not with placebo (ES = 0.06, p = 0.8; between-group: ES = 0.68, p = 0.05). Curcumin also improved SRT Total (ES = 0.53, p = 0.002), visual memory (BVMT-R Recall: ES = 0.50, p = 0.01; BVMT-R Delay: ES = 0.51, p = 0.006), and attention (ES = 0.96, p < 0.0001) compared with placebo (ES = 0.28, p = 0.1; between-group: ES = 0.67, p = 0.04). FDDNP binding decreased significantly in the amygdala with curcumin (ES = −0.41, p = 0.04) compared with placebo (ES = 0.08, p = 0.6; between-group: ES = 0.48, p = 0.07). In the hypothalamus, FDDNP binding did not change with curcumin (ES = −0.30, p = 0.2), but increased with placebo (ES = 0.26, p = 0.05; between-group: ES = 0.55, p = 0.02). Conclusions: Daily oral Theracurmin may lead to improved memory and attention in non-demented adults. The FDDNP-PET findings suggest that symptom benefits are associated with decreases in amyloid and tau accumulation in brain regions modulating mood and memory.

Autosomal dominant Alzheimer's disease (ADAD) is a small subset of Alzheimer's disease that is genetically determined with 100% penetrance. It provides a valuable window into studying the course of pathologic processes that leads to dementia. Arterial spin labeling (ASL) MRI is a potential AD imaging marker that non-invasively measures cerebral perfusion. In this study, we investigated the relationship of cerebral blood flow measured by pseudo-continuous ASL (pCASL) MRI with measures of cerebral metabolism (FDG PET) and amyloid deposition (Pittsburgh Compound B (PiB) PET). Thirty-one participants at risk for ADAD (age 39 ± 13 years, 19 females) were recruited into this study, and 21 of them received both MRI and FDG and PiB PET scans. Considerable variability was observed in regional correlations between ASL-CBF and FDG across subjects. Both regional hypo-perfusion and hypo-metabolism were associated with amyloid deposition. Cross-sectional analyses of each biomarker as a function of the estimated years to expected dementia diagnosis indicated an inverse relationship of both perfusion and glucose metabolism with amyloid deposition during AD development. These findings indicate that neurovascular dysfunction is associated with amyloid pathology, and also indicate that ASL CBF may serve as a sensitive early biomarker for AD. The direct comparison among the three biomarkers provides complementary information for understanding the pathophysiological process of AD., Highlights • Regional associations of three imaging biomarkers (pCASL-MRI, FDG-PET and PiB-PET) are studied in persons at risk for ADAD. • PiB-PET shows different spatial pattern compared to pCASL-MRI and FDG-PET. • There is considerable variability among regional correlations between pCASL and FDG. • Both regional hypo-perfusion and hypo-metabolism are associated with amyloid deposition.

Currently, only presumptive diagnosis of chronic traumatic encephalopathy (CTE) can be made in living patients. We present a modality that may be instrumental to the definitive diagnosis of CTE in living patients based on brain autopsy confirmation of [F-18]FDDNP-PET findings in an American football player with CTE. [F-18]FDDNP-PET imaging was performed 52 mo before the subject's death. Relative distribution volume parametric images and binding values were determined for cortical and subcortical regions of interest. Upon death, the brain was examined to identify the topographic distribution of neurodegenerative changes. Correlation between neuropathology and [F-18]FDDNP-PET binding patterns was performed using Spearman rank-order correlation. Mood, behavioral, motor, and cognitive changes were consistent with chronic traumatic myeloencephalopathy with a 22-yr lifetime risk exposure to American football. There were tau, amyloid, and TDP-43 neuropathological substrates in the brain with a differential topographically selective distribution. [F-18]FDDNP-PET binding levels correlated with brain tau deposition (r(s) = 0.59, P = .02), with highest relative distribution volumes in the parasagittal and paraventricular regions of the brain and the brain stem. No correlation with amyloid or TDP-43 deposition was observed. [F-18]FDDNP-PET signals may be consistent with neuropathological patterns of tau deposition in CTE, involving areas that receive the maximal shearing, angular–rotational acceleration–deceleration forces in American football players, consistent with distinctive and differential topographic vulnerability and selectivity of CTE beyond brain cortices, also involving midbrain and limbic areas. Future studies are warranted to determine whether differential and selective [F-18]FDDNP-PET may be useful in establishing a diagnosis of CTE in at-risk patients.

Logan graphical analysis with cerebellum as reference region has been widely used for the estimation of the distribution volume ratio (DVR) of [18F]FDDNP as a measure of amyloid burden and tau deposition in human brain because of its simplicity and computational ease. However, spurious parametric DVR images may be produced with shorter scanning times and when the noise level is high. In this work, we have characterized a relative-equilibrium-based (RE) graphical method against the Logan analysis for parametric imaging and region-of-interest (ROI) analysis. Methods Dynamic [18F]FDDNP PET scans were performed on 9 control subjects and 12 patients diagnosed with Alzheimer's disease. Using the cerebellum as reference input, regional DVR estimates were derived using both the Logan analysis and the RE plot approach. Effects on DVR estimates obtained at voxel and ROI levels by both graphical approaches using data in different time windows were investigated and compared with the standard values derived using the Logan analysis on a voxel-by-voxel basis for the time window of 35–125 min used in previous studies. Results Larger bias and variability were observed for DVR estimates obtained by the Logan graphical analysis at the voxel level when short time windows (85–125 and 45–65 min) were used, because of high noise levels in voxel-wise parametric imaging. However, when the Logan graphical analysis was applied at the ROI level over those short time windows, the DVR estimates did not differ significantly from the standard values derived using the Logan analysis on the voxel level for the time window of 35–125 min, and their bias and variability were remarkably lower. Conversely, the RE plot approach was more robust in providing DVR estimates with less bias and variability even when short time windows were used. The DVR estimates obtained at voxel and ROI levels were consistent. No significant differences were observed in DVR estimates obtained by the RE plot approach for all paired comparisons with the standard values. Conclusions The RE plot approach provides less noisy parametric images and gives consistent and reliable regional DVR estimates at both voxel and ROI levels, indicating that it is preferred over the Logan graphical analysis for analyzing [18F]FDDNP PET data.

The effects of dietary condition and blood glucose level on the kinetics and uptake of (18)F-FDG in mice were systematically investigated using intraperitoneal and tail-vein injection.Dynamic PET was performed for 60 min on 23 isoflurane-anesthetized male C57BL/6 mice after intravenous (n = 11) or intraperitoneal (n = 12) injection of (18)F-FDG. Five and 6 mice in the intravenous and intraperitoneal groups, respectively, were kept fasting overnight (18 ± 2 h), and the others were fed ad libitum. Serial blood samples were collected from the femoral artery to measure (18)F-FDG and glucose concentrations. Image data were reconstructed using filtered backprojection with CT-based attenuation correction. The standardized uptake value (SUV) was estimated from the 45- to 60-min image. The metabolic rate of glucose (MRGlu) and (18)F-FDG uptake constant (K(i)) were derived by Patlak graphical analysis.In the brain, SUV and K(i) were significantly higher in fasting mice with intraperitoneal injection, but MRGlu did not differ significantly under different dietary states and administration routes. Cerebral K(i) was inversely related to elevated blood glucose levels, irrespective of administration route or dietary state. In myocardium, SUV, K(i), and MRGlu were significantly lower in fasting than in nonfasting mice for both routes of injection. Myocardial SUV and K(i) were strongly dependent on the dietary state, and K(i) did not correlate with the blood glucose level. Similar results were obtained for skeletal muscle, although the differences were not as pronounced.Intraperitoneal injection is a valid alternative route, providing pharmacokinetic data equivalent to data from tail-vein injection for small-animal (18)F-FDG PET. Cerebral K(i) varies inversely with blood glucose level, but the measured cerebral MRGlu does not correlate with blood glucose level or dietary condition. Conversely, the K(i) values of the myocardium and skeletal muscle are strongly dependent on dietary condition but not on blood glucose level. In tissue in which (18)F-FDG uptake declines with increasing blood glucose, correction for blood glucose level will make SUV a more robust outcome measure of MRGlu.

Glycosylphosphatidylinositol-anchored high density lipoprotein-binding protein 1 (GPIHBP1), a GPI-anchored endothelial cell protein, binds lipoprotein lipase (LPL) and transports it into the lumen of capillaries where it hydrolyzes triglycerides in lipoproteins. GPIHBP1 is assumed to be expressed mainly within the heart, skeletal muscle, and adipose tissue, the sites where most lipolysis occurs, but the tissue pattern of GPIHBP1 expression has never been evaluated systematically. Because GPIHBP1 is found on the luminal face of capillaries, we predicted that it would be possible to define GPIHBP1 expression patterns with radiolabeled GPIHBP1-specific antibodies and positron emission tomography (PET) scanning. In Gpihbp1(-/-) mice, GPIHBP1-specific antibodies were cleared slowly from the blood, and PET imaging showed retention of the antibodies in the blood pools (heart and great vessels). In Gpihbp1(+/+) mice, the antibodies were cleared extremely rapidly from the blood and, to our surprise, were taken up mainly by lung and liver. Immunofluorescence microscopy confirmed the presence of GPIHBP1 in the capillary endothelium of both lung and liver. In most tissues with high levels of Gpihbp1 expression, Lpl expression was also high, but the lung was an exception (very high Gpihbp1 expression and extremely low Lpl expression). Despite low Lpl transcript levels, however, LPL protein was readily detectable in the lung, suggesting that some of that LPL originates elsewhere and then is captured by GPIHBP1 in the lung. In support of this concept, lung LPL levels were significantly lower in Gpihbp1(-/-) mice than in Gpihbp1(+/+) mice. In addition, Lpl(-/-) mice expressing human LPL exclusively in muscle contained high levels of human LPL in the lung.

© 2015, National Academy of Sciences. All rights reserved. Chronic traumatic encephalopathy (CTE) is an acquired primary tauopathy with a variety of cognitive, behavioral, and motor symptoms linked to cumulative brain damage sustained from single, episodic, or repetitive traumatic brain injury (TBI). No definitive clinical diagnosis for this condition exists. In this work, we used [F-18]FDDNP PET to detect brain patterns of neuropathology distribution in retired professional American football players with suspected CTE (n = 14) and compared results with those of cognitively intact controls (n = 28) and patients with Alzheimer' s dementia (AD) ( n = 24), a disease that has been cognitively associated with CTE. [F-18]FDDNP PET imaging results in the retired players suggested the presence of neuropathological patterns consistent with models of concussion wherein brainstem white matter tracts undergo early axonal damage and cumulative axonal injuries along subcortical, limbic, and cortical brain circuitries supporting mood, emotions, and behavior. This deposition pattern is distinctively different from the progressive pattern of neuropathology [paired helical filament (PHF)-tau and amyloid-β] in AD, which typically begins in the medial temporal lobe progressing along the cortical default mode network, with no or minimal involvement of subcortical structures. This particular [F-18]FDDNP PET imaging pattern in cases of suspected CTE also is primarily consistent with PHF-tau distribution observed at autopsy in subjects with a history of mild TBI and autopsy-confirmed diagnosis of CTE.

Medical parametric imaging with dynamic positron emission tomography (PET) plays an increasingly potential role in modern biomedical research and clinical diagnosis. The key issue in parametric imaging is to estimate parameters based on sampled data at the pixel-by-pixel level from certain dynamic processes described by valid mathematical models. Classic nonlinear least squares (NLS) algorithm requires a ''good'' initial guess and the computational time-complexity is high, which is impractical for image-wide parameter estimation. Although a variety of fast parametric imaging techniques have been developed, most of them focus on single input systems, which do not provide an optimal solution for dual-input biomedical system parameter estimation, which is the case of liver metabolism. In this study, a dual-input-generalized linear least squares (D-I-GLLS) algorithm was proposed to identify the model parameters including the parameter in the dual-input function. Monte Carlo simulation was conducted to examine this novel fast algorithm. The results of the quantitative analysis suggested that the proposed technique could provide comparable reliability of the parameter estimation with NLS fitting and accurately identify the parameter in the dual-input function. This method may be potentially applicable to other dual-input biomedical system parameter estimation as well.

Dynamic SPECT has the potential to provide absolute physiological parameter estimates. However, the low sensitivity of SPECT typically results in very noisy dynamic SPECT data. Filtering can reduce the noise, but at the expense of degrading the already poor resolution further. The effect of reconstruction parameters, post-reconstruction filtering and resolution recovery on kinetic parameter estimation bias and reliability was systematically investigated. Dynamic projection data were generated using Monte Carlo (MC) simulations of a mathematical brain phantom at 10 different levels of Poisson noise. The projection data were reconstructed with OSEM with varying numbers of iterations and subsets and were filtered with three-dimensional (3-D) Gaussian filters with varying FWHM. Bias and reliability of the main parameters of interest (K/sub 1/,V/sub d/, and binding potential) for thalamus, cerebellum, and frontal cortex were estimated for the three-compartment model fits to the tissue time-activity curves derived from the reconstructed data. Reliability (standard deviation) of parameter estimates was obtained with the Bootstrap MC technique, which showed good agreement with conventional MC in a subset of data sets, but required only a small fraction of conventional MC computation time. Post-reconstruction filtering increased bias, without improving the reliability of parameter estimates and, hence, no post-reconstruction filtering is recommended. For reconstructions without resolution recovery, an effective number of 40 iterations overall provided the best tradeoff between bias and reliability for the range of noise levels studied. Resolution recovery achieved a modest reduction in bias.

Accurate determination of the input function is essential for absolute quantification of physiological parameters in positron emission tomography and single-photon emission computed tomography imaging, but it requires an invasive and tedious procedure of blood sampling that is impractical in clinical studies. We previously proposed a technique that estimates simultaneously kinetic parameters and the input function from the tissue impulse response functions and requires two blood samples. A nonlinear least squares method estimated all the parameters in the impulse response functions and the input function but failed occasionally due to high noise levels in the data, causing an ill-conditioned cost function. This paper investigates the feasibility of applying a Monte Carlo method called simulated annealing to estimate kinetic parameters in the impulse response functions and the input function. Time-activity curves of teboroxime, which is very sensitive to changes in the input function, were simulated based on published data obtained from a canine model. The equations describing the tracer kinetics in different regions were minimized simultaneously by simulated annealing and nonlinear least squares. We found that the physiological parameters obtained with simulated annealing are accurate, and the estimated input function more closely resembled the simulated curve. We conclude that simulated annealing reduces bias in the estimation of physiological parameters and determination of the input function.

Quantitative positron emission tomography (PET) studies provide in vivo measurements of dynamic physiological and biochemical processes in humans. A limitation of PET is an inability to provide precise anatomic localization due to relatively poor spatial resolution when compared to magnetic resonance (MR) imaging. Manual placement of region-of-interest (ROI) is commonly used in clinical and research settings in analysis of PET datasets. However, this approach is operator dependent and time-consuming. A semi- or fully-automated ROI delineation (or segmentation) method offers advantages by reducing operator error/subjectivity and thereby improving reproducibility. In this work, we describe an approach to automatically segment dynamic PET images using cluster analysis and we validate our approach with a simulated phantom study and assess its performance with real dynamic PET data. Our preliminary results suggest that cluster analysis can automatically segment tissues in dynamic PET studies and has the potential to replace manual ROI delineation for some applications.

5-iodo-3-(2(S)-azetidinylmethoxy)pyridine (5-[123I]iodo-A-85380) is a novel radioligand for brain nicotinic acetylcholine receptors (nAChRs), having demonstrated high specific binding to nAChRs and favourable properties for in vivo imaging. The purpose of this study was to evaluate different tracer kinetic modelling strategies for the quantification of regional nAChR binding in baboon brain using dynamic single-photon emission computed tomography (SPECT). Imaging studies were performed on two anaesthetised baboons. The radioligand (397 ± 43 MBq) was infused into the animals over 2 min and dynamic SPECT images were acquired for 3 hr. The plasma input function was determined from arterial blood samples with metabolite correction. The data were analysed using 2- and 3-compartment models. Reliable fits could be obtained with a 3-compartment model which provided significantly improved fitting as compared to the 2-compartment model, despite poor identifiability in some parameters. The 2-compartment model failed to fit the data adequately even though the identifiability of parameters was high. The 3-compartment model described the data well, and with the strategies used to obtain reliable fits, is the model of choice for analysis of 5-[123I]iodo-A-85380 kinetics.

Head movement during a dynamic brain PET/CT imaging results in mismatch between CT and dynamic PET images. It can cause artifacts in CT-based attenuation corrected PET images, thus affecting both the qualitative and quantitative aspects of the dynamic PET images and the derived parametric images. In this study, we developed an automated retrospective image-based movement correction (MC) procedure. The MC method first registered the CT image to each dynamic PET frames, then re-reconstructed the PET frames with CT-based attenuation correction, and finally re-aligned all the PET frames to the same position. We evaluated the MC method's performance on the Hoffman phantom and dynamic FDDNP and FDG PET/CT images of patients with neurodegenerative disease or with poor compliance. Dynamic FDDNP PET/CT images (65 min) were obtained from 12 patients and dynamic FDG PET/CT images (60 min) were obtained from 6 patients. Logan analysis with cerebellum as the reference region was used to generate regional distribution volume ratio (DVR) for FDDNP scan before and after MC. For FDG studies, the image derived input function was used to generate parametric image of FDG uptake constant (Ki) before and after MC. Phantom study showed high accuracy of registration between PET and CT and improved PET images after MC. In patient study, head movement was observed in all subjects, especially in late PET frames with an average displacement of 6.92 mm. The z-direction translation (average maximum = 5.32 mm) and x-axis rotation (average maximum = 5.19 degrees) occurred most frequently. Image artifacts were significantly diminished after MC. There were significant differences (P

Coded apertures provide a substantial gain in detection efficiency compared with conventional collimation and are well suited to imaging small volumes. Here, the authors investigated several aspects of coded aperture design for a small animal SPECT system, including aperture/detector configuration, sampling requirements, and susceptibility to scatter. They simulated various source distributions and detection systems which included one, two, and four stationary detectors placed around the object, each with a pinhole array or a Fresnel zone plate in front of the detector. Image volumes were reconstructed using an iterative successive over-relaxation algorithm with a penalized weighted least squares cost function. Multiple pinhole arrays performed better than Fresnel zone plates in terms of reconstructed mean squared error and signal-to-noise. The authors' design goals of 1% detection efficiency can be achieved with a four-detector system with arrays of 100 pinholes per detector and the scatter fraction for a 2-cm diameter object is

Optimal sampling schedule (OSS) is of great interest in biomedical experiment design, as it can improve the physiological parameter estimation precision and significantly reduce the samples required. A number of well designed algorithms and software packages have been developed, which deal with the instantaneous measurements at discrete times. However, in nuclear medicine tracer kinetic studies, the imaging systems, such as positron emission tomography (PET) and single photon emission computed tomography (SPECT), take measurements (images) based on continuous accumulation over time intervals. In this case, the existing algorithms cannot be used to design OSS so as to reduce the image frame numbers. In this paper, a general OSS design algorithm for the accumulative measurement is proposed. The potential usefulness of the algorithm is demonstrated by its designing OSS in [18F] fluoro-2-deoxy- d -glucose (FDG) studies with PET to estimate the local cerebral metabolic rate of glucose. The robustness of parameter estimation using the OSS with respect to intra-subject and inter-subject parameter variations is also presented.

Dynamic imaging with positron emission tomography (PET) is widely used for the in vivo measurement of regional cerebral metabolic rate for glucose (rCMRGlc) with [18F]fluorodeoxy-D-glucose (FDG) and is used for the clinical evaluation of neurological disease. However, in addition to the acquisition of dynamic images, continuous arterial blood sampling is the conventional method to obtain the tracer time-activity curve in blood (or plasma) for the numeric estimation of rCMRGlc in mg glucose/100-g tissue/min. The insertion of arterial lines and the subsequent collection and processing of multiple blood samples are impractical for clinical PET studies because it is invasive, has the remote, but real potential for producing limb ischemia, and it exposes personnel to additional radiation and risks associated with handling blood. In this paper, based on our previously proposed method for extracting kinetic parameters from dynamic PET images, we developed a modified version (post-estimation method) to improve the numerical identifiability of the parameter estimates when we deal with data obtained from clinical studies. We applied both methods to dynamic neurologic FDG PET studies in three adults. We found that the input function and parameter estimates obtained with our noninvasive methods agreed well with those estimated from the gold standard method of arterial blood sampling and that rCMRGlc estimates were highly correlated (r = 0.973). More importantly, no significant difference was found between rCMRGlc estimated by our methods and the gold standard method (P0.16). We suggest that our proposed noninvasive methods may offer an advance over existing methods.

Positron emission tomography (PET) is an important tool for enabling quantification of human brain function. However, quantitative studies using tracer kinetic modeling require the measurement of the tracer time-activity curve in plasma (PTAC) as the model input function. It is widely believed that the insertion of arterial lines and the subsequent collection and processing of the biomedical signal sampled from the arterial blood are not compatible with the practice of clinical PET, as it is invasive and exposes personnel to the risks associated with the handling of patient blood and radiation dose. Therefore, it is of interest to develop practical noninvasive measurement techniques for tracer kinetic modeling with PET. In this paper, a technique is proposed to extract the input function together with the physiological parameters from the brain dynamic images alone. The identifiability of this method is tested rigorously by using Monte Carlo simulation. The results show that the proposed method is able to quantify all the required parameters by using the information obtained from two or more regions of interest (ROI's) with very different dynamics in the PET dynamic images. There is no significant improvement in parameter estimation for the local cerebral metabolic rate of glucose (LCMRGlc) if the number of ROI's are more than three. The proposed method can provide very reliable estimation of LCMRGlc, which is our primary interest in this study.

Accurate determination of the plasma input function (IF) is essential for absolute quantification of physiological parameters in positron emission tomography (PET). However, it requires an invasive and tedious procedure of arterial blood sampling that is challenging in mice because of the limited blood volume. In this study, a hybrid modeling approach is proposed to estimate the plasma IF of 2-deoxy-2-[18F]fluoro-D-glucose ([18F]FDG) in mice using accumulated radioactivity in urinary bladder together with a single late-time blood sample measurement.Dynamic PET scans were performed on nine isoflurane-anesthetized male C57BL/6 mice after a bolus injection of [18F]FDG at the lateral caudal vein. During a 60- or 90-min scan, serial blood samples were taken from the femoral artery. Image data were reconstructed using filtered backprojection with computed tomography-based attenuation correction. Total accumulated radioactivity in the urinary bladder at late times was fitted to a renal compartmental model with the last blood sample and a one-exponential function that described the [18F]FDG clearance in blood. Multiple late-time blood sample estimates were calculated by the blood [18F]FDG clearance equation. A sum of four-exponentials was assumed for the plasma IF that served as a forcing function to all tissues. The estimated plasma IF was obtained by simultaneously fitting the [18F]FDG model to the time-activity curves (TACs) of liver and muscle and the forcing function to early (0-1 min) left-ventricle data (corrected for delay, dispersion, partial-volume effects, and erythrocyte uptake) and the late-time blood estimates. Using only the blood sample collected at the end of the study to estimate the IF and the use of liver TAC as an alternative IF were also investigated.The area under the plasma IFs calculated for all studies using the hybrid approach was not significantly different from that using all blood samples. [18F]FDG uptake constants in brain, myocardium, skeletal muscle, and liver computed by the Patlak analysis using estimated and measured plasma IFs were in excellent agreement (slope∼1; R20.983). The IF estimated using only the last blood sample drawn at the end of the study and the use of liver TAC as the plasma IF provided less reliable results.The estimated plasma IFs obtained with the hybrid method agreed well with those derived from arterial blood sampling. Importantly, the proposed method obviates the need of arterial catheterization, making it possible to perform repeated dynamic [18F]FDG PET studies on the same animal. Liver TAC is unsuitable as an input function for absolute quantification of [18F]FDG PET data.

This work is aimed at correlating pre-mortem [18F]FDDNP positron emission tomography (PET) scan results in a patient with dementia with Lewy bodies (DLB), with cortical neuropathology distribution determined postmortem in three physical dimensions in whole brain coronal sections. Analysis of total amyloid-β (Aβ) distribution in frontal cortex and posterior cingulate gyrus confirmed its statistically significant correlation with cortical [18F]FDDNP PET binding values (distribution volume ratios, DVR) (p < 0.001, R = 0.97, R2 = 0.94). Neurofibrillary tangle (NFT) distribution correlated significantly with cortical [18F]FDDNP PET DVR in the temporal lobe (p < 0.001, R = 0.87, R2 = 0.76). Linear combination of Aβ and NFT densities was highly predictive of [18F]FDDNP PET DVR through all analyzed regions of interest (p < 0.0001, R = 0.92, R2 = 0.85), and both densities contributed significantly to the model. Lewy bodies were present at a much lower level than either Aβ or NFTs and did not significantly contribute to the in vivo signal. [18F]FDG PET scan results in this patient were consistent with the distinctive DLB pattern of hypometabolism. This work offers a mapping brain model applicable to all imaging probes for verification of imaging results with Aβ and/or tau neuropathology brain distribution using immunohistochemistry, fluorescence microscopy, and autoradiography.

Intra-scan body movement (IBM) during a PET/CT study can degrade the image quality in subtle ways that are not easy for physicians to detect and can thus affect the diagnostic value of the images. In this study, we propose a method for identifying regions on a PET/CT image that are affected by IBM. The method is based on bootstrapping the sinogram data of short sub-frames of a scan to form multiple sets of sinograms of original scan length, from which multiple images are reconstructed. High voxel-wise percent standard variation among these images would indicate image regions that are compromised by IBM and should be interpreted with caution. Both computer simulation and real human PET/CT data were used in this study to demonstrate the validity of the proposed method for identifying IBM.

In this chapter, the present status and future possibilities for computer-aided-detection (CAD) in breast imaging is considered. Xeromammography and later, conventional x-ray mammography, were among the earliest medical imaging modalities to benefit from the use of computers to assist radiologists in detecting lesions, especially cancers. At present, mammography is the preferred method to screen asymptomatic women for breast cancer. Breast cancer itself is a heterogeneous disease with no cure; the earlier the cancer is detected, the better the prognosis. The principal thrust of CAD research in recent years has therefore been to detect early signs of the disease, such as microcalcifications, small masses, and subtle lesions, especially those most likely to be missed by radiologists, so that any cancer present may be detected at the earliest possible stage of the disease. In this chapter, CAD in breast imaging is reviewed, and the possible lines of future research and development speculated on. The major unresolved problems are identified and, in some cases, promising trends and possible solutions are outlined. Mammography has certain structural deficiencies that have propelled research into alternative imaging modalities for breast cancer detection. Some of these emerging imaging modalities that could either be adjuncts to mammography or supplant it in the future are reviewed. The possible roles for CAD for these alternative modalities are also examined. Certain generic problems of CAD, such as accurate segmentation, registration, lesion detection, and assessment of algorithm performance are then considered. The technology of CAD, in the context of mammography, generally stands for computer-aided detection of lesions and suspicious regions, meriting careful scrutiny by a radiologist. If a patient's history and the radiologist's findings are taken into account, together with the computer-aided detection data that provides diagnostic output, a computer-aided diagnosis (CADx) system exists. Sometimes, a computer-aided diagnosis system is also confusingly referred to by the acronym CAD. In an attempt to overcome such confusion, a computer-aided detection system is sometimes referred to as a CADe system. In this chapter, computer-aided detection is referred to as CAD, and computer-aided diagnosis as CADx.

Many tumor cells have elevated rates of glucose uptake that can be measured quantitatively, noninvasively and repeatedly by positron emission tomography (PET) with 2-deoxy-2-[ 18 F]-fluoro-D-glucose ( 18 F-FDG). Clinical imaging with 18 F-FDG PET has been used for detection and staging of primary and metastatic tumors. High-resolution microPET scanning and murine cancer models make it possible to analyze longitudinally glucose metabolism during the appearance, development and progression of individual experimental tumors. In this study, we used 18 F-FDG microPET and micro computerized tomography (microCT) to investigate glucose uptake in the DMBA/TPA chemically-induced multistage mouse skin carcinogenesis model. 18 F-FDG uptake is significantly higher in all papillomas than in surrounding skin. Elevated 18 F-FDG uptake is observed when tumors can be identified morphologically, but not before. Although 18 F-FDG uptake is high in all fully invasive, malignant skin squamous cell carcinomas, uptake in papillomas and microinvasive malignant squamous cell carcinomas is variable and does not exhibit any correlation with tumor stage.

Positron emission tomography (PET) with 2-(1-{6-[(2-[18F]fluoroethyl)(methyl)amino]-2-naphthyl}ethylidene)malo-nonitrile ([18F]FDDNP) has been used for in vivo imaging of β-amyloid plaques and neurofibrillary tangles, the neuropathological hallmarks of Alzheimer's disease (AD). Due to its simplicity as compared to other reference tissue modeling methods, Logan graphical analysis with cerebellum as reference region has been primarily used to estimate the distribution volume ratio (DVR) of [18F]FDDNP as a measure of amyloid burden and tau deposition in human brain. However, it sometimes produces spurious parametric DVR images when the noise level is high and the total scan duration is shortened. The purpose of this study was to characterize a new graphical analysis method against the Logan analysis for the estimation of the DVR estimates at regional and voxel-wise levels, using the cerebellum as the reference region. Effects on regional (frontal and medial temporal) DVR estimates and parametric DVR images obtained by the Logan and the new approaches using data in different time intervals and total scan durations (35–125 min, 65–125 min, 85–125 min, and 45–65 min) were investigated and compared to the reference values derived by the Logan analysis for the time interval 35–125 min. For regional analyses, with the exception of DVR estimates of medial temporal in ADs obtained by the Logan analysis for the interval 85–125 min, no significant difference was observed for all paired comparisons with the reference values. Logan DVR images generated using shortened time intervals (85–125 min and 45–65 min) were extremely noisy. In contrast, the new method produced DVR images comparable to the reference images even when shortened time intervals were used. The new method provides more reliable regional DVR estimates and less noisy parametric images, indicating that it is an improvement over the Logan graphical analysis for analyzing [18F]FDDNP PET data.

Accurate determination of the input function is essential for absolute quantification of physiological parameters in PET. However, it requires an invasive and tedious procedure of arterial blood sampling that is challenging in mice because of the limited blood volume and the small size of the blood vessels. Blood loss and surgical trauma may limit the same mouse to be studied longitudinally. In this paper, we proposed a modeling approach to estimate whole-blood sample measurements for 2-[18F]fluoro-2-deoxy-D-glucose ([18F]FDG) PET studies in mice by use of urinary bladder time-activity data and one blood sample taken at the end of the experiment. Six CL57BL/6 mice were injected with [18F]FDG, and serial blood samples were taken from the femoral artery via a surgically implanted catheter during small-animal PET scans. Image data were reconstructed by use of filtered backprojection with CT-based attenuation correction. The boundary of the urinary bladder was determined with the fused PET-CT image by adjusting the scaling of the PET image such that the bladder volume shown on the PET and CT images matched. The bladder time-activity data were then constructed frame-by-frame and fitted to a compartmental model in conjunction with a blood sample acquired near the end of the experiment and an one-exponential function that described the clearance of [18F]FDG in blood. Our results demonstrated that it is feasible to estimate [18F]FDG blood samples by use of urinary bladder time-activity data. The estimated samples obtained with the proposed method agreed well with those derived from the gold standard method of arterial blood sampling. More importantly, the proposed method obviates the need of arterial catheterization, making it possible to perform repeated studies on the same mouse.

Detection, localization, diagnosis, staging, and monitoring treatment responses are the most important aspects and crucial procedures in diagnostic medicine and clinical oncology. Early detection and localization of the diseases and accurate disease staging can improve the survival and change management in patients prior to planned surgery or therapy. Therefore, current medical practice has been directed toward early but efficient localization and staging of diseases, while ensuring that patients would receive the most effective treatment.

2-(l-{6-[(2-[18F]fluoroethyl)(methyl)amino]-2-naph- thyl}ethylidene)malononitrile ([18F]FDDNP) PET studies for imaging neuronal concentration of beta-amyloid plaques and neurofibrillary tangles, the neuropathological hallmarks of Alzheimer's disease (AD), usually requires a 1-2 hour dynamic scan to measure the kinetics of the tracer in brain tissue. It is difficult for the subject to keep his or her head stationary during the entire data acquisition period. Thus, the reliability of the acquired kinetic data could be severely compromised. Furthermore, misalignment between transmission (TX) and emission (EM) scans due to head movement could introduce errors in attenuation correction (AC) and physiological parameters. In this study, we have developed an image-based procedure for head movement correction that takes a TX image and coregisters it to its corresponding non-AC EM frame before reconstruction of the dynamic image and then realigns all the now proper AC EM frames to a common reference frame. Effects of head movement on tracer time-activity curves and distribution volume ratio (DVR) parametric images are also investigated. The proposed movement correction procedure is done with [18F]FDDNP as a case in point but has the potential to be used in a wide variety of dynamic brain PET studies.

Positron emission tomography (PET) with 2-(1-{6- [(2-[18F]fluoroethyl)(methyl)amino]-2-naphthyl}ethylidene)malo-nonitrile ([18F]FDDNP) has been used for in vivo imaging of amyloid plaques and neurofibrillary tangles, the neuopathological hallmarks of Alzheimer's disease (AD). The purpose of this study was to assess seven noninvasive methods for the estimation of the distribution volume ratio (DVR), as a measure of binding density of [18F]FDDNP in human brain, using a reference input (cerebellum). Imaging studies were performed on 20 AD and 13 normal control subjects. The image data were analyzed using two reference tissue modeling approaches with and without a physiological constraint on the efflux rate constant from reference tissue, Logan graphical analysis with and without a population efflux rate constant from reference tissue, and a simple concentration ratio between target and reference regions, to derive the DVR parameter. Highly linear correlations were observed among the DVR estimates except those by the tissue ratio approach which varied considerably. Reference tissue modeling approaches provided reliable estimation of DVR and gave further insight in the delivery of [18F]FDDNP to regions known to be affected by amyloid deposition. The efflux rate constant of cerebellum had minor effect on the DVR derived by Logan analysis. The [18F]FDDNP retention in subcortical white matter was similar between controls and ADs. Our results suggest that (1) Logan graphical analysis is the most robust method for quantitative analysis of [18F]FDDNP PET data without blood sampling, (2) reference tissue modeling approaches can provide further insight in helping to understand the transport properties of [18F]FDDNP and to aid model construction for kinetic analysis using the arterial input function, (3) efflux rate constant of the reference tissue can be derived with higher reliability by model fitting with physiological constraint, and (4) white matter may potentially be used as a reference region for analyzing [18F]FDDNP PET data.

The last few decades of the twentieth century have witnessed significant advances in medical imaging, which had been exclusively concerned with conventional film/screen X-ray imaging for more than 75 years after the discovery of X-ray by Wilhelm Rontgen in 1895. In particular, when Allen Cormack and Godfrey Hounsfield introduced X-ray computed tomography (CT) independently in the early 1970s [1–3] based on the mathematical foundation laid by Radon [4] for reconstructing images of an object from its projections, the field of medical imaging was revolutionized. Imaging with X-ray CT has enabled us to view noninvasively, for the first time, the anatomic structure of internal organs with unprecedentedprecision and to recognize any gross pathology of organs anddiseases. This alsomarked a newera ofmedical diagnosticswithmany invasive and potentially morbid procedures being substituted by noninvasive cross-sectional imaging. The breakthrough development of X-ray CT was made possible by continuing advances in instrumentation and computer technologies, which also accelerated the development of other multi-dimensional imaging modalities that possess a great potential for providing, in addition to morphologic (structural)

Accurate determination of the input function is essential for absolute quantification of physiological parameters in positron emission tomography (PET) and single-photon emission computed tomography (SPECT) imaging. Uncertainties associated with the input function due to such factors as measurement errors and inappropriate sampling protocols can hamper reliable estimation of physiological parameters. Although intravenous bolus injection has been a widely used tracer administration schedule, it imposes a sampling requirement around the peak of the blood curve where rapid blood sampling is needed to capture the dramatic changes of tracer kinetics. By administering tracer as a short-time constant infusion, such a sampling requirement can be relieved. Fitting an analytical function or a series of functions to the blood data is a common strategy to reduce statistical fluctuations. However, abrupt changes in the blood curve and differed in the injection schedule can cause ill-conditioned fitting. A physiological model that incorporated the injection schedule should be more appropriate for this task. The focus of this paper is to evaluate one such model for fitting blood data obtained with a short-time infusion of [18F] fluorodeoxyglucose (FDG). It was found that common analytical functions failed to fit the blood data in majority cases, whereas the physiological model that incorporated the injection schedule provided reliable and reproducible fits to the blood data.

With the increasing use of teleradiology systems, large amount of data is acquired and transmitted, thus raising the issue of medical image compression. The goal of image compression is to represent an image with as few number of bits as possible while preserving the quality required for the given application. Standard lossless compression schemes can only yield compression ratios of about 2:1 that is insufficient to compress volumetric tomographic image data. Common standards available in industry for lossy image compression are usually used in non-medical applications and their application to medical image compression is still under investigation. Fractal image coding is a new compression technique that has received much attention recently. This study investigates the feasibility of applying fractal image coding approach to nuclear medicine image compression. A quadtree-based partition scheme was used in the encoding phase to partition an image frame into small irregular segments that after decoding process, yield an approximate image to the original. Our preliminary results show that compression ratios higher than 10:1 can be achieved in clinical images. It was also found that there is no diagnostic loss in the parametric images computed from the reconstructed images as compared to those obtained from the original raw data. We conclude that fractal image coding could be used to compress tomographic images and it may be useful in telemedicine.

Accurate determination of the input function is essential for absolute quantification of physiological parameters in PET and SPECT imaging but it requires an invasive and tedious procedure of blood sampling that is impractical in clinical studies. We previously proposed a technique that simultaneously estimates kinetic parameters and the input function from the tissue impulse response functions and which requires only two blood samples. A nonlinear least squares method was used to estimate all the parameters in the impulse response functions and the input function but it fails occasionally due to high noise levels in the data causing an ill-conditioned cost function. This study investigates the feasibility of applying a Monte Carlo method called simulated annealing to estimate kinetic parameters in the impulse response functions and the input function. Time-activity curves of teboroxime, which is very sensitive to changes in the input function, were simulated based on published data obtained from a canine model. The equations describing the tracer kinetics in different regions were minimised simultaneously by simulated annealing and nonlinear least squares. We found that the physiological parameters obtained with simulated annealing are more accurate and the estimated input function more closely resembled the simulated curve. We conclude that simulated annealing reduces bias in the estimation of physiological parameters and determination of the input function.

The authors investigated the biomechanical properties of the plantar foot tissue in diabetic patients using an ultrasound indentation system. Four neuropathic diabetes mellitus (DM) patients and four normal subjects were chosen for this study. Indentation tests were performed at four sites (the 1st metatarsal head, 2nd metatarsal head, big toe and heel) in selected postures. The tissue thickness and effective Young's modulus were measured by the ultrasound indentation system. Statistic analyses showed that the Young's moduli of DM patients were significantly higher than those of normal subjects at various sites. Meanwhile, the tissue thicknesses of DM patients were significantly (p