Comparison of 3-D reconstruction with 3D-OSEM and with FORE+OSEM for PET

3D-OSEM versus FORE + OSEM: Optimal Reconstruction Algorithm for FDG PET with a Short Acquisition Time

Objective  In this study, we investigated the optimal reconstruction algorithm in fluorodeoxyglucose (FDG) positron emission tomography (PET) with a short acquisition time. Materials and Methods  In the phantom study, six spheres filled with FDG solution (sphere size: 6.23-37 mm; radioactivity ratio of spheres to background = 8:1) and placed in a National Electrical Manufacturers Association phantom were evaluated. Image acquisition time was 15 to 180 seconds, and the obtained image data were reconstructed using each of the Fourier rebinning (FORE) + ordered subsets expectation-maximization (OSEM) and 3D-OSEM algorithms. In the clinical study, mid-abdominal images of 19 patients were evaluated using regions of interest placed on areas of low, intermediate, and high radioactivity. All obtained images were investigated visually, and quantitatively using maximum standardized uptake value (SUV) and coefficient of variation (CV). Results  In the phantom study, FORE + OSEM images with a short acquisition time had large CVs (poor image quality) but comparatively constant maximum SUVs. 3D-OSEM images showed comparatively constant CVs (good image quality) but significantly low maximum SUVs. The results of visual evaluation were well correlated with those of quantitative evaluation. Small spheres were obscured on 3D-OSEM images with short acquisition time, but image quality was not greatly deteriorated. The clinical and phantom studies yielded similar results. Conclusion  FDG PET images with a short acquisition time reconstructed by FORE + OSEM showed poorer image quality than by 3D-OSEM. However, images obtained with a short acquisition time and reconstructed with FORE + OSEM showed clearer FDG uptake and more useful than 3D-OSEM in the light of the detection of lesions.

Comparison of ordered-subset expectation maximization and filtered back projection reconstruction based on quantitative outcome from dynamic [18F]NaF PET images

[18F]NaF PET imaging is a useful tool for measuring regional bone metabolism. However, due to tracer in urine, [18F]NaF PET images of the hip reconstructed using filtered back projection (FBP) frequently show streaking artifacts in slices through the bladder leading to noisy time-activity curves unsuitable for quantification. This study compares differences between quantitative outcomes at the hip derived from images reconstructed using the FBP and ordered-subset expectation maximization (OSEM) methods. Dynamic [18F]NaF PET data at the hip for four postmenopausal women were reconstructed using FBP and nine variations of the OSEM algorithm (all combinations of 1, 5, 15 iterations and 10, 15, 21 subsets). Seven volumes of interest were placed in the hip. Bone metabolism was measured using standardized uptake values, Patlak analysis (Ki-PAT) and Hawkins model Ki-4k. Percentage differences between the standardized uptake values and Ki values from FBP and OSEM images were assessed. OSEM images appeared visually smoother and without the streaking artifacts seen with FBP. However, due to loss of counts, they failed to recover the quantitative values in VOIs close to the bladder, including the femoral head and femoral neck. This was consistent for all quantification methods. Volumes of interest farther from the bladder or larger and receiving greater counts showed good convergence with 5 iterations and 21 subsets. For VOIs close to the bladder, including the femoral neck and femoral head, 15 iterations and 10, 15 or 21 subsets were not enough to obtain OSEM images suitable for measuring bone metabolism and showed no improvement compared to FBP.

Bias evaluation and reduction in 3D OP-OSEM reconstruction in dynamic equilibrium PET studies with 11C-labeled for binding potential analysis

Iterative image reconstruction is widely used in positron emission tomography. However, it is known to contribute to quantitation bias and is particularly pronounced during dynamic studies with ¹¹C-labeled radiotracers where count rates become low towards the end of the acquisition. As the strength of the quantitation bias depends on the counts in the reconstructed frame, it can differ from frame to frame of the acquisition. This is especially relevant in the case of neuro-receptor studies with simultaneous PET/MR when a bolus-infusion protocol is applied to allow the comparison of pre- and post-task effects. Here, count dependent changes in quantitation bias may interfere with task changes. We evaluated the impact of different framing schemes on quantitation bias and its propagation into binding potential (BP) using a phantom decay study with ¹¹C and 3D OP-OSEM. Further, we propose a framing scheme that keeps the true counts per frame constant over the acquisition time as constant framing schemes and conventional increasing framing schemes are unlikely to achieve stable bias values during the acquisition time range. For a constant framing scheme with 5 minutes frames, the BP bias was 7.13±2.01% (10.8% to 3.8%) compared to 5.63±2.85% (7.8% to 4.0%) for conventional increasing framing schemes. Using the proposed constant true counts framing scheme, a stabilization of the BP bias was achieved at 2.56±3.92% (3.5% to 1.7%). The change in BP bias was further studied by evaluating the linear slope during the acquisition time interval. The lowest slope values were observed in the constant true counts framing scheme. The constant true counts framing scheme was effective for BP bias stabilization at relevant activity and time ranges. The mean BP bias under these conditions was 2.56±3.92%, which represents the lower limit for the detection of changes in BP during equilibrium and is especially important in the case of cognitive tasks where the expected changes are low.

APRICOT: Advanced Platform for Reproducible Infrastructures in the Cloud via Open Tools

Background  Scientific publications are meant to exchange knowledge among researchers but the inability to properly reproduce computational experiments limits the quality of scientific research. Furthermore, bibliography shows that irreproducible preclinical research exceeds 50%, which produces a huge waste of resources on nonprofitable research at Life Sciences field. As a consequence, scientific reproducibility is being fostered to promote Open Science through open databases and software tools that are typically deployed on existing computational resources. However, some computational experiments require complex virtual infrastructures, such as elastic clusters of PCs, that can be dynamically provided from multiple clouds. Obtaining these infrastructures requires not only an infrastructure provider, but also advanced knowledge in the cloud computing field.

Objectives  The main aim of this paper is to improve reproducibility in life sciences to produce better and more cost-effective research. For that purpose, our intention is to simplify the infrastructure usage and deployment for researchers.

Methods  This paper introduces Advanced Platform for Reproducible Infrastructures in the Cloud via Open Tools (APRICOT), an open source extension for Jupyter to deploy deterministic virtual infrastructures across multiclouds for reproducible scientific computational experiments. To exemplify its utilization and how APRICOT can improve the reproduction of experiments with complex computation requirements, two examples in the field of life sciences are provided. All requirements to reproduce both experiments are disclosed within APRICOT and, therefore, can be reproduced by the users.

Results  To show the capabilities of APRICOT, we have processed a real magnetic resonance image to accurately characterize a prostate cancer using a Message Passing Interface cluster deployed automatically with APRICOT. In addition, the second example shows how APRICOT scales the deployed infrastructure, according to the workload, using a batch cluster. This example consists of a multiparametric study of a positron emission tomography image reconstruction.

Conclusion  APRICOT's benefits are the integration of specific infrastructure deployment, the management and usage for Open Science, making experiments that involve specific computational infrastructures reproducible. All the experiment steps and details can be documented at the same Jupyter notebook which includes infrastructure specifications, data storage, experimentation execution, results gathering, and infrastructure termination. Thus, distributing the experimentation notebook and needed data should be enough to reproduce the experiment.

Optimization-Based Image Reconstruction From Low-Count, List-Mode TOF-PET Data

OBJECTIVE

We investigate an optimization-based approach to image reconstruction from list-mode data in digital time-of-flight (TOF) positron emission tomography (PET) imaging.

METHOD

In the study, the image to be reconstructed is designed as a solution to a convex, non-smooth optimization program, and a primal-dual algorithm is developed for image reconstruction by solving the optimization program. The algorithm is first applied to list-mode TOF-PET data of a typical count level from physical phantoms and a human subject. Subsequently, we explore the algorithm's potential for image reconstruction in low-dose and/or fast TOF-PET imaging of practical interest by applying the algorithm to list-mode TOF-PET data of different, low-count levels from the same physical phantoms and human subject.

RESULTS

Visual inspection and quantitative-metric analysis reveal that the optimization reconstruction approach investigated can yield images with enhanced spatial and contrast resolution, suppressed image noise, and increased axial volume coverage over the reference images obtained with a standard clinical reconstruction algorithm especially for low-dose TOF-PET data.

SIGNIFICANCE

The optimization-based reconstruction approach can be exploited for yielding insights into potential quality upper bound of reconstructed images in, and design of scanning protocols of, TOF-PET imaging of practical significance.

History and future technical innovation in positron emission tomography

Instrumentation for positron emission tomography (PET) imaging has experienced tremendous improvements in performance over the past 60 years since it was first conceived as a medical imaging modality. Spatial resolution has improved by a factor of 10 and sensitivity by a factor of 40 from the early designs in the 1970s to the high-performance scanners of today. Multimodality configurations have emerged that combine PET with computed tomography (CT) and, more recently, with MR. Whole-body scans for clinical purposes can now be acquired in under 10 min on a state-of-the-art PET/CT. This paper will review the history of these technical developments over 40 years and summarize the important clinical research and healthcare applications that have been made possible by these technical advances. Some perspectives for the future of this technology will also be presented that promise to bring about new applications of this imaging modality in clinical research and healthcare.

4-Dimensional MRI and Attenuation Map Generation in PET/MRI with 4-Dimensional PET-Derived Deformation Matrices: Study of Feasibility for Lung Cancer Applications

Respiratory motion may reduce accuracy in the fusion of functional and anatomic images from combined PET/MRI systems. Methodologies for the correction of respiratory motion in PET acquisitions with such systems are mostly based on the use of respiration-synchronized MRI acquisitions to derive motion fields. Existing approaches based on tagging acquisitions may introduce artifacts in MR images, whereas motion model approaches require the acquisition of training datasets. The objective of this work was to investigate the possibility of generating 4-dimensional (4D) MR images and associated attenuation maps (AMs) from the combination of a single static MR image and motion fields obtained from simultaneously acquired 4D non-attenuation-corrected (NAC) PET images. Methods: Four-dimensional PET/MRI datasets were acquired for 11 patients on a simultaneous PET/MRI system. The 4D PET datasets were retrospectively binned into 4 motion amplitude frames corresponding to the simultaneously acquired T1-weighted 4D MR images. A T1-weighted 3-dimensional MRI sequence with Dixon-based fat and water separation was also acquired at the end of expiration for PET attenuation correction purposes. All reconstructed 4D NAC PET images were then elastically registered to the single end-of-expiration NAC PET image. The derived motion fields were subsequently applied to the end-of-expiration frame of the acquired 4D MRI volume and the AM derived from the Dixon MR image to generate respiration-synchronized MR images and corresponding AMs. Results: The accuracy of the proposed method was assessed by comparing the generated and acquired images according to metrics such as overall correlation coefficients and differences in distances of anatomic landmarks on the generated and acquired MRI datasets. High correlation coefficients (mean ± SD: 0.93 ± 0.03) and small differences (2.69 ± 0.5 mm) were obtained. Moreover, small tissue classification differences (2.23% ± 0.68%) between generated and 4D MRI-extracted AMs were observed. Conclusion: Our results confirm the feasibility of using 4D NAC PET images for accurate PET attenuation correction and respiratory motion correction in PET/MRI, without the need for patient-specific 4D MRI acquisitions.

Investigation of optimization-based reconstruction with an image-total-variation constraint in PET

Interest remains in reconstruction-algorithm research and development for possible improvement of image quality in current PET imaging and for enabling innovative PET systems to enhance existing, and facilitate new, preclinical and clinical applications. Optimization-based image reconstruction has been demonstrated in recent years of potential utility for CT imaging applications. In this work, we investigate tailoring the optimization-based techniques to image reconstruction for PET systems with standard and non-standard scan configurations. Specifically, given an image-total-variation (TV) constraint, we investigated how the selection of different data divergences and associated parameters impacts the optimization-based reconstruction of PET images. The reconstruction robustness was explored also with respect to different data conditions and activity up-takes of practical relevance. A study was conducted particularly for image reconstruction from data collected by use of a PET configuration with sparsely populated detectors. Overall, the study demonstrates the robustness of the TV-constrained, optimization-based reconstruction for considerably different data conditions in PET imaging, as well as its potential to enable PET configurations with reduced numbers of detectors. Insights gained in the study may be exploited for developing algorithms for PET-image reconstruction and for enabling PET-configuration design of practical usefulness in preclinical and clinical applications.

In vivo imaging of neuromelanin in Parkinson's disease using 18F-AV-1451 PET

The tau tangle ligand (18)F-AV-1451 ((18)F-T807) binds to neuromelanin in the midbrain, and may therefore be a measure of the pigmented dopaminergic neuronal count in the substantia nigra. Parkinson's disease is characterized by progressive loss of dopaminergic neurons. Extrapolation of post-mortem data predicts that a ∼30% decline of nigral dopamine neurons is necessary to cause motor symptoms in Parkinson's disease. Putamen dopamine terminal loss at disease onset most likely exceeds that of the nigral cell bodies and has been estimated to be of the order of 50-70%. We investigated the utility of (18)F-AV-1451 positron emission tomography to visualize the concentration of nigral neuromelanin in Parkinson's disease and correlated the findings to dopamine transporter density, measured by (123)I-FP-CIT single photon emission computed tomography. A total of 17 patients with idiopathic Parkinson's disease and 16 age- and sex-matched control subjects had (18)F-AV-1451 positron emission tomography using a Siemens high-resolution research tomograph. Twelve patients with Parkinson's disease also received a standardized (123)I-FP-CIT single photon emission computed tomography scan at our imaging facility. Many of the patients with Parkinson's disease displayed visually apparent decreased (18)F-AV-1451 signal in the midbrain. On quantitation, patients showed a 30% mean decrease in total nigral (18)F-AV-1451 volume of distribution compared with controls (P = 0.004), but there was an overlap of the individual ranges. We saw no significant correlation between symptom dominant side and contralateral nigral volume of distribution. There was no correlation between nigral (18)F-AV-1451 volume of distribution and age or time since diagnosis. In the subset of 12 patients, who also had a (123)I-FP-CIT scan, the mean total striatal dopamine transporter signal was decreased by 45% and the mean total (18)F-AV-1451 substantia nigra volume of distribution was decreased by 33% after median disease duration of 4.7 years (0.5-12.4 years). (18)F-AV-1451 positron emission tomography may be the first radiotracer to reflect the loss of pigmented neurons in the substantia nigra of parkinsonian patients. The magnitude of the nigral signal loss was smaller than the decrease in striatal dopamine transporter signal measured by dopamine transporter single photon emission computed tomography. These findings suggest a more severe loss of striatal nerve terminal function compared with neuronal cell bodies, in accordance with the post-mortem literature.

MAP reconstruction for Fourier rebinned TOF-PET data

Time-of-flight (TOF) information improves the signal-to-noise ratio in positron emission tomography (PET). The computation cost in processing TOF-PET sinograms is substantially higher than for nonTOF data because the data in each line of response is divided among multiple TOF bins. This additional cost has motivated research into methods for rebinning TOF data into lower dimensional representations that exploit redundancies inherent in TOF data. We have previously developed approximate Fourier methods that rebin TOF data into either three-dimensional (3D) nonTOF or 2D nonTOF formats. We refer to these methods respectively as FORET-3D and FORET-2D. Here we describe maximum a posteriori (MAP) estimators for use with FORET rebinned data. We first derive approximate expressions for the variance of the rebinned data. We then use these results to rescale the data so that the variance and mean are approximately equal allowing us to use the Poisson likelihood model for MAP reconstruction. MAP reconstruction from these rebinned data uses a system matrix in which the detector response model accounts for the effects of rebinning. Using these methods we compare the performance of FORET-2D and 3D with TOF and nonTOF reconstructions using phantom and clinical data. Our phantom results show a small loss in contrast recovery at matched noise levels using FORET compared to reconstruction from the original TOF data. Clinical examples show FORET images that are qualitatively similar to those obtained from the original TOF-PET data but with a small increase in variance at matched resolution. Reconstruction time is reduced by a factor of 5 and 30 using FORET3D+MAP and FORET2D+MAP respectively compared to 3D TOF MAP, which makes these methods attractive for clinical applications.

Effect of varying number of OSEM subsets on PET lesion detectability

Iterative reconstruction has become the standard for routine clinical PET imaging. However, iterative reconstruction is computationally expensive, especially for time-of-flight (TOF) data. Block-iterative algorithms such as ordered-subsets expectation maximization (OSEM) are commonly used to accelerate the reconstruction. There is a tradeoff between the number of subsets and reconstructed image quality. The objective of this work was to evaluate the effect of varying the number of OSEM subsets on lesion detection for general oncologic PET imaging.

METHODS

Experimental phantom data were taken from the Utah PET Lesion Detection Database, modeling whole-body oncologic (18)F-FDG PET imaging of a 92-kg patient. The experiment consisted of 24 scans over 4 d on a TOF PET/CT scanner, with up to 23 lesions (diameter, 6-16 mm) distributed throughout the thorax, abdomen, and pelvis. Images were reconstructed with maximum-likelihood expectation maximization (MLEM) and with OSEM using 2-84 subsets. The reconstructions were repeated both with and without TOF. Localization receiver-operating-characteristic (LROC) analysis was applied using the channelized nonprewhitened observer. The observer was first used to optimize the number of iterations and smoothing filter for each case that maximized lesion-detection performance for these data; this was done to ensure that fair comparisons were made with each test case operating near its optimal performance. The probability of correct localization and the area under the LROC curve were then analyzed as functions of the number of subsets to characterize the effect of OSEM on lesion-detection performance.

RESULTS

Compared with the baseline MLEM algorithm, lesion-detection performance with OSEM declined as the number of subsets increased. The decline was moderate out to about 12-14 subsets and then became progressively steeper as the number of subsets increased. Comparing TOF with non-TOF results, the magnitude of the performance drop was larger for TOF reconstructions.

CONCLUSION

PET lesion-detection performance is degraded when OSEM is used with a large number of subsets. This loss of image quality can be controlled using a moderate number of subsets (e.g., 12-14 or fewer), retaining a large degree of acceleration while maintaining high image quality. The use of more aggressive subsetting can result in image quality degradations that offset the benefits of using TOF or longer scan times.

PET∕CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain

PURPOSE

Transcranial focused ultrasound (FUS) delivers highly focused acoustic energy to a small region of the brain in a noninvasive manner. Recent studies have revealed that FUS, which is administered either in pulsed or continuous waves, can elicit or suppress neural tissue excitability. This neuromodulatory property of FUS has been demonstrated via direct motion detection, electrophysiological recordings, functional magnetic resonance imaging (fMRI), confocal imaging, and microdialysis sampling of neurotransmitters. This study presents new evidence of local increase in glucose metabolism induced by FUS to the rat brain using FDG (18-fludeoxyglucose) positron emission tomography (PET).

METHODS

Sprague-Dawley rats underwent sonication to a unilateral hemispheric area of the brain prior to PET scan. The pulsed sonication (350 kHz, tone burst duration of 0.5 ms, pulse repetition frequency of 1 kHz, and duration of 300 ms) was applied in 2 s intervals for 40 min immediately after the FDG injection via tail vein. Subsequently, the PET was acquired in dynamic list-mode to image FDG activity for an hour, and reconstructed into a single volume representing standardized uptake value (SUV). The raw SUV as well as its asymmetry index (AI) were measured from five different volume-of-interests (VOIs) of the brain for both hemispheres, and compared between sonicated and unsonicated groups.

RESULTS

Statistically significant hemispheric changes in SUV were observed only at the center of sonication focus within the FUS group [paired t-test; t(7) = 3.57, p < 0.05]. There were no significant hemispheric differences in SUV within the control group in any of the VOIs. A statistically significant elevation in AI (t-test; t(7) = 3.40, p < 0.05) was observed at the center of sonication focus (7.9 ± 2.5%, the deviations are in standard error) among the FUS group when compared to the control group (-0.8 ± 1.2%).

CONCLUSIONS

Spatially distinct increases in the glucose metabolic activity in the rat brain is present only at the center of sonication focus, suggesting localized functional neuromodulation mediated by the sonication.

3D PET image reconstruction including both motion correction and registration directly into an MR or stereotaxic spatial atlas

This work explores the feasibility and impact of including both the motion correction and the image registration transformation parameters from positron emission tomography (PET) image space to magnetic resonance (MR), or stereotaxic, image space within the system matrix of PET image reconstruction. This approach is motivated by the fields of neuroscience and psychiatry, where PET is used to investigate differences in activation patterns between different groups of participants, requiring all images to be registered to a common spatial atlas. Currently, image registration is performed after image reconstruction which introduces interpolation effects into the final image. Furthermore, motion correction (also requiring registration) introduces a further level of interpolation, and the overall result of these operations can lead to resolution degradation and possibly artifacts. It is important to note that performing such operations on a post-reconstruction basis means, strictly speaking, that the final images are not ones which maximize the desired objective function (e.g. maximum likelihood (ML), or maximum a posteriori reconstruction (MAP)). To correctly seek parameter estimates in the desired spatial atlas which are in accordance with the chosen reconstruction objective function, it is necessary to include the transformation parameters for both motion correction and registration within the system modeling stage of image reconstruction. Such an approach not only respects the statistically chosen objective function (e.g. ML or MAP), but furthermore should serve to reduce the interpolation effects. To evaluate the proposed method, this work investigates registration (including motion correction) using 2D and 3D simulations based on the high resolution research tomograph (HRRT) PET scanner geometry, with and without resolution modeling, using the ML expectation maximization (MLEM) reconstruction algorithm. The quality of reconstruction was assessed using bias-variance and root mean squared error analyses, comparing the proposed method to conventional post-reconstruction registration methods. An overall reduction in bias (for a cold region: from 41% down to 31% (2D) and 97% down to 65% (3D), and for a hot region: from 11% down to 8% (2D) and from 16% down to 14% (3D)) and in root mean squared error analyses (for a cold region: from 43% to 37% (2D) and from 97% to 65% (3D), and for a hot region: from 11% to 9% (2D) and from 16% down to 14% (3D)) in reconstructed regional mean activities (full regions of interest; all with statistical significance: p < 5 × 10(-10)) is found when including the motion correction and registration in the system matrix of the MLEM reconstruction, with resolution modeling. However, this improvement in performance comes with an extra computational cost of about 40 min. In this context, this work constitutes an important step toward the goal of estimating parameters of interest directly from the raw Poisson-distributed PET data, and hence toward the complete elimination of post-processing steps.

Quantitative accuracy of MAP reconstruction for dynamic PET imaging in small animals

PURPOSE

Iterative reconstruction algorithms are becoming more commonly employed in positron emission tomography (PET) imaging; however, the quantitative accuracy of the reconstructed images still requires validation for various levels of contrast and counting statistics.

METHODS

The authors present an evaluation of the quantitative accuracy of the 3D maximum a posteriori (3D-MAP) image reconstruction algorithm for dynamic PET imaging with comparisons to two of the most widely used reconstruction algorithms: the 2D filtered-backprojection (2D-FBP) and 2D-ordered subsets expectation maximization (2D-OSEM) on the Siemens microPET scanners. The study was performed for various levels of count density encountered in typical dynamic scanning as well as the imaging of cardiac activity concentration in small animal studies on the Focus 120. Specially designed phantoms were used for evaluation of the spatial resolution, image quality, and quantitative accuracy. A normal mouse was employed to evaluate the accuracy of the blood time activity concentration extracted from left ventricle regions of interest (ROIs) within the images as compared to the actual blood activity concentration measured from arterial blood sampling.

RESULTS

For MAP reconstructions, the spatial resolution and contrast have been found to reach a stable value after 20 iterations independent of the β values (i.e., hyper parameter which controls the weight of the penalty term) and count density within the frame. The spatial resolution obtained with 3D-MAP reaches values of ∼1.0 mm with a β of 0.01 while the 2D-FBP has value of 1.8 mm and 2D-OSEM has a value of 1.6 mm. It has been observed that the lower the hyper parameter β used in MAP, more iterations are needed to reach the stable noise level (i.e., image roughness). The spatial resolution is improved by using a lower β value at the expense of higher image noise. However, with similar noise level the spatial resolution achieved by 3D-MAP was observed to be better than that by 2D-FBP or 2D-OSEM. Using an image quality phantom containing hot spheres, the estimated activity concentration in the largest sphere has the expected concentration relative to the background area for all the MAP images. The obtained recovery coefficients have been also shown to be almost independent of the count density. 2D-FBP and 2D-OSEM do not perform as well, yielding recovery coefficients lower than those observed with 3D-MAP (approximately 33% lower for the smallest sphere). However, a small positive bias was observed in MAP reconstructed images for frames of very low count density. This bias is present in the uniform area for count density of less than 0.05 × 10(6) counts/ml. For the dynamic mouse study, it was observed that 3D-MAP (even gated at diastole) cannot predict accurately the blood activity concentration due to residual spill-over activity from the myocardium into the left ventricle (approximately 15%). However, 3D-MAP predicts blood activity concentration closer to blood sampling than 2D-FBP.

CONCLUSIONS

The authors observed that 3D-MAP produces more accurate activity concentration estimates than 2D-FBP or 2D-OSEM at all practical levels of statistics and contrasts due to improved spatial resolution leading to lesser partial volume effect.

Glucose metabolism in small subcortical structures in Parkinson's disease

OBJECTIVES

Evidence from experimental animal models of Parkinson's disease (PD) suggests a characteristic pattern of metabolic perturbation in discrete, very small basal ganglia structures. These structures are generally too small to allow valid investigation by conventional positron emission tomography (PET) cameras. However, the high-resolution research tomograph (HRRT) PET system has a resolution of 2 mm, sufficient for the investigation of important structures such as the pallidum and thalamic subnuclei.

MATERIALS AND METHODS

Using the HRRT, we performed [(18)F]-fluorodeoxyglucose (FDG) scans on 21 patients with PD and 11 age-matched controls. We employed three types of normalization: white matter, global mean, and data-driven normalization. We performed volume-of-interest analyses of small subcortical gray matter structures. Voxel-based comparisons were performed to investigate the extent of cortical hypometabolism.

RESULTS

The most significant level of relative subcortical hypermetabolism was detected in the external pallidum (GPe), irrespective of normalization strategy. Hypermetabolism was suggested also in the internal pallidum, thalamic subnuclei, and the putamen. Widespread cortical hypometabolism was seen in a pattern very similar to previously reported patterns in patients with PD.

CONCLUSION

The presence and extent of subcortical hypermetabolism in PD is dependent on type of normalization. However, the present findings suggest that PD, in addition to widespread cortical hypometabolism, is probably characterized by true hypermetabolism in the GPe. This finding was predicted by the animal 2-deoxyglucose autoradiography literature, in which high-magnitude hypermetabolism was also most robustly detected in the GPe.

Optimal rebinning of time-of-flight PET data

Time-of-flight (TOF) positron emission tomography (PET) scanners offer the potential for significantly improved signal-to-noise ratio (SNR) and lesion detectability in clinical PET. However, fully 3D TOF PET image reconstruction is a challenging task due to the huge data size. One solution to this problem is to rebin TOF data into a lower dimensional format. We have recently developed Fourier rebinning methods for mapping TOF data into non-TOF formats that retain substantial SNR advantages relative to sinograms acquired without TOF information. However, mappings for rebinning into non-TOF formats are not unique and optimization of rebinning methods has not been widely investigated. In this paper we address the question of optimal rebinning in order to make full use of TOF information. We focus on FORET-3D, which approximately rebins 3D TOF data into 3D non-TOF sinogram formats without requiring a Fourier transform in the axial direction. We optimize the weighting for FORET-3D to minimize the variance, resulting in H(2)-weighted FORET-3D, which turns out to be the best linear unbiased estimator (BLUE) under reasonable approximations and furthermore the uniformly minimum variance unbiased (UMVU) estimator under Gaussian noise assumptions. This implies that any information loss due to optimal rebinning is as a result only of the approximations used in deriving the rebinning equation and developing the optimal weighting. We demonstrate using simulated and real phantom TOF data that the optimal rebinning method achieves variance reduction and contrast recovery improvement compared to nonoptimized rebinning weightings. In our preliminary study using a simplified simulation setup, the performance of the optimal rebinning method was comparable to that of fully 3D TOF MAP.

Biomedical signal and image processing

Generally, physiological modeling and biomedical signal processing constitute two important paradigms of biomedical engineering (BME): their fundamental concepts are taught starting from undergraduate studies and are more completely dealt with in the last years of graduate curricula, as well as in Ph.D. courses. Traditionally, these two cultural aspects were separated, with the first one more oriented to physiological issues and how to model them and the second one more dedicated to the development of processing tools or algorithms to enhance useful information from clinical data. A practical consequence was that those who did models did not do signal processing and vice versa. However, in recent years,the need for closer integration between signal processing and modeling of the relevant biological systems emerged very clearly [1], [2]. This is not only true for training purposes(i.e., to properly prepare the new professional members of BME) but also for the development of newly conceived research projects in which the integration between biomedical signal and image processing (BSIP) and modeling plays a crucial role. Just to give simple examples, topics such as brain–computer machine or interfaces,neuroengineering, nonlinear dynamical analysis of the cardiovascular (CV) system,integration of sensory-motor characteristics aimed at the building of advanced prostheses and rehabilitation tools, and wearable devices for vital sign monitoring and others do require an intelligent fusion of modeling and signal processing competences that are certainly peculiar of our discipline of BME.

Physical and clinical performance of the mCT time-of-flight PET/CT scanner

Time-of-flight (TOF) measurement capability promises to improve PET image quality. We characterized the physical and clinical PET performance of the first Biograph mCT TOF PET/CT scanner (Siemens Medical Solutions USA, Inc.) in comparison with its predecessor, the Biograph TruePoint TrueV. In particular, we defined the improvements with TOF. The physical performance was evaluated according to the National Electrical Manufacturers Association (NEMA) NU 2-2007 standard with additional measurements to specifically address the TOF capability. Patient data were analyzed to obtain the clinical performance of the scanner. As expected for the same size crystal detectors, a similar spatial resolution was measured on the mCT as on the TruePoint TrueV. The mCT demonstrated modestly higher sensitivity (increase by 19.7 ± 2.8%) and peak noise equivalent count rate (NECR) (increase by 15.5 ± 5.7%) with similar scatter fractions. The energy, time and spatial resolutions for a varying single count rate of up to 55 Mcps resulted in 11.5 ± 0.2% (FWHM), 527.5 ± 4.9 ps (FWHM) and 4.1 ± 0.0 mm (FWHM), respectively. With the addition of TOF, the mCT also produced substantially higher image contrast recovery and signal-to-noise ratios in a clinically-relevant phantom geometry. The benefits of TOF were clearly demonstrated in representative patient images.

Fully-3D PET image reconstruction using scanner-independent, adaptive projection data and highly rotation-symmetric voxel assemblies

For iterative, fully 3D positron emission tomography (PET) image reconstruction intrinsic symmetries can be used to significantly reduce the size of the system matrix. The precalculation and beneficial memory-resident storage of all nonzero system matrix elements is possible where sufficient compression exists. Thus, reconstruction times can be minimized independently of the used projector and more elaborate weighting schemes, e.g., volume-of-intersection (VOI), are applicable. A novel organization of scanner-independent, adaptive 3D projection data is presented which can be advantageously combined with highly rotation-symmetric voxel assemblies. In this way, significant system matrix compression is achieved. Applications taking into account all physical lines-of-response (LORs) with individual VOI projectors are presented for the Siemens ECAT HR+ whole-body scanner and the Siemens BrainPET, the PET component of a novel hybrid-MR/PET imaging system. Measured and simulated data were reconstructed using the new method with ordered-subset-expectation-maximization (OSEM). Results are compared to those obtained by the sinogram-based OSEM reconstruction provided by the manufacturer. The higher computational effort due to the more accurate image space sampling provides significantly improved images in terms of resolution and noise.

Hybrid imaging technology: from dreams and vision to clinical devices

Early in the history of nuclear medicine imaging it was realized that the nature of physiological mechanisms associated with the use of radiotracers prevented the identification of anatomic structures with a high degree of accuracy. This limitation often created difficulties in accurate interpretations of acquired images and caused investigators to seek methods of obtaining accurate anatomic correlations. Initial work centered on the use of software tools to combine anatomic and physiological data. Limitations in the use of these techniques, coupled with the development and refinements of anatomic imaging technologies (computed tomography [CT] and magnetic resonance imaging [MRI]), resulted in the development of hybrid imaging systems that combined CT with single-photon emission computed tomography (SPECT) and positron emission tomography (PET). With these hybrid systems, the images can be viewed separately or combined in a fused presentation for direct image correlation of anatomy and physiology. Presently, SPECT systems are available either with nondiagnostic CT capability for attenuation correction and image correlation, or with fully diagnostic CT capability, providing complementary diagnostic information. Equivalently, PET systems with diagnostic CT capability that provide high-resolution physiological and anatomic images are also now commercially available. These systems continue to evolve with the development of new detector materials and data acquisition and image processing technology. The widespread use of SPECT in cardiac imaging has resulted in the development of several new approaches to data acquisition and these new systems currently have either CT capability or the addition of this technology is planned in the future. The development and commercial availability of hybrid imaging systems has provided physicians with important new tools that significantly improve the diagnostic, staging, and treatment planning processes that are now available for their use.

Quantification of regional myocardial blood flow in a canine model of stunned and infarcted myocardium: comparison of rubidium-82 positron emission tomography with microspheres

BACKGROUND

Myocardial viability and quantification of regional myocardial blood flow (MBF) are important for the diagnosis of heart disease. Positron emission tomography is the current gold standard for determining myocardial viability, but most positron-emitting perfusion tracers require an on-site cyclotron. Rubidium-82 ((82)Rb) is a myocardial perfusion tracer that is produced using an on-site generator. This study investigates (82)Rb-measured MBF in canine models of stunned and infarcted myocardium compared with selected measurements obtained concurrently using microspheres.

METHODS

Myocardial stunning and infarction were created in canines by occluding the left anterior descending for 15 min and 2 h, respectively. Stunning was produced in all animals; six animals were reperfused after the 2 h occlusion, whereas the other six animals remained occluded permanently. Regional MBF was measured in each group during rest and dobutamine stress at acute and chronic (8 weeks postinsult) time points using dynamic (82)Rb perfusion imaging and radioactively labeled microspheres.

RESULTS

Average resting MBF with microspheres and Rb was 0.68+/-0.02 versus 0.73+/-0.01 (P<0.001) in nonischemic tissue, and 0.53+/-0.03 versus 0.42+/-0.02 (P<0.001) in the region-at-risk tissue, respectively. Average MBF during stress with microspheres and Rb was 2.78+/-0.15 versus 3.53+/-0.16 (P<0.05) in the nonischemic tissue, and 1.90+/-0.20 versus 2.31+/-0.26 (P = NS) in the region-at-risk tissue, respectively.

CONCLUSION

Despite the small significant differences, the dynamic (82)Rb measurements provide estimates of MBF in stunned and acutely and chronically infarcted tissue at rest and during hyperemia that correspond with clinical interpretation.

3D versus 2D dynamic 82Rb myocardial blood flow imaging in a canine model of stunned and infarcted myocardium

PURPOSE

Previous studies have shown the ability of rubidium-82 ((82)Rb) positron emission tomography (PET) imaging to quantitatively measure myocardial blood flow (MBF), many of which are performed using two-dimensional (2D) imaging. Three-dimensional (3D) imaging provides increased sensitivity and may result in decreased costs owing to a reduction in the required injected activity of radiotracer. This study compares 2D and 3D (82)Rb PET MBF results obtained in the same imaging session.

METHODS

Three-dimensional and 2D (82)Rb perfusion imaging was performed in canines on a GE Discovery LS PET/CT scanner at rest and during hyperemia in stunned and infarcted tissue. MBF (ml/min/g) was determined using a 1-compartment model and an extraction correction of the uptake rate and analyzed using a standard 17-segment model.

RESULTS

A strong, significant correlation was present (rho = 0.95, P<0.0001). Average 3D MBF values were slightly lower at rest and higher during stress versus 2D. MBF results in normal, stunned, and infarcted tissue differed by 7% on average and significant increases in MBF from rest to hyperemia were noted with both the techniques.

CONCLUSION

These results imply that MBF results obtained in 3D are comparable with traditional 2D imaging. Therefore, it may be possible to use 3D imaging with lower administered activity, helping to reduce costs and patient dose without compromising quantitative information.

[Effects of transmission scan protocol and attenuation correction method on normal database of 2-[18F]fluoro-2-deoxy-D-glucose (18F-FDG) brain positron emission tomography study]

Although post-injection transmission scan (POST-TS) after 2-[(18)F]fluoro-2-deoxy- D-glucose ((18)F-FDG) injection[A1] is useful for short examination times, the emission count of (18)F-FDG[A2] in the regional brain area was not completely subtracted with use of the POST-TS method. The purpose of this study was to investigate the effect of POST-TS and attenuation correction (AC) methods on the normal database (NDB). A 10 min pre-injection transmission scan (PRE-TS) was performed before (18)F-FDG[A3] was injected in eighteen normal volunteers. A 10 min POST-TS was then conducted beginning 40 min after (18)F-FDG[A4] injection, followed by a 10 min 2-dimentional emission scanning. To reconstruct each image of normal volunteers, the reconstruction was performed using the filtered back-projection (FBP) method and the ordered subsets expectation maximization (OSEM) method, with transmission-based measured attenuation correction (MAC) and the segmented attenuation correction (SAC) technique. Subtraction images of NDB with PRE-TS or POST-TS were evaluated using 3D-SSP. A phantom study was also performed in addition to a human study, and assessment was by region of interests and profile curves. NDB images with POST-TS were significantly lower in the bilateral frontal lobes and higher in the parietal lobes and occipital lobes, including the precuneus, than those with PRE-TS, regardless of the different AC and reconstruction algorithms. Therefore, we have to be careful to confirm not only emission scan methods and reconstruction algorithms, but also TS methods and AC methods in the NDB. It will be best to perform PET examinations using the same TS methods and AC methods between NDB and patients.

Lesion quantification in oncological positron emission tomography: a maximum likelihood partial volume correction strategy

A maximum likelihood (ML) partial volume effect correction (PVEC) strategy for the quantification of uptake and volume of oncological lesions in 18F-FDG positron emission tomography is proposed. The algorithm is based on the application of ML reconstruction on volumetric regional basis functions initially defined on a smooth standard clinical image and iteratively updated in terms of their activity and volume. The volume of interest (VOI) containing a previously detected region is segmented by a k-means algorithm in three regions: A central region surrounded by a partial volume region and a spill-out region. All volume outside the VOI (background with all other structures) is handled as a unique basis function and therefore "frozen" in the reconstruction process except for a gain coefficient. The coefficients of the regional basis functions are iteratively estimated with an attenuation-weighted ordered subset expectation maximization (AWOSEM) algorithm in which a 3D, anisotropic, space variant model of point spread function (PSF) is included for resolution recovery. The reconstruction-segmentation process is iterated until convergence; at each iteration, segmentation is performed on the reconstructed image blurred by the system PSF in order to update the partial volume and spill-out regions. The developed PVEC strategy was tested on sphere phantom studies with activity contrasts of 7.5 and 4 and compared to a conventional recovery coefficient method. Improved volume and activity estimates were obtained with low computational costs, thanks to blur recovery and to a better local approximation to ML convergence.

Experimental comparison of lesion detectability for four fully-3D PET reconstruction schemes

The objective of this work was to evaluate the lesion detection performance of four fully-3D positron emission tomography (PET) reconstruction schemes using experimentally acquired data. A multi-compartment anthropomorphic phantom was set up to mimic whole-body (18)F-fluorodeoxyglucose (FDG) cancer imaging and scanned 12 times in 3D mode, obtaining count levels typical of noisy clinical scans. Eight of the scans had 26 (68)Ge "shell-less" lesions (6, 8-, 10-, 12-, 16-mm diameter) placed throughout the phantom with various target:background ratios. This provided lesion-present and lesion-absent datasets with known truth appropriate for evaluating lesion detectability by localization receiver operating characteristic (LROC) methods. Four reconstruction schemes were studied: 1) Fourier rebinning (FORE) followed by 2D attenuation-weighted ordered-subsets expectation-maximization, 2) fully-3D AW-OSEM, 3) fully-3D ordinary-Poisson line-of-response (LOR-)OSEM; and 4) fully-3D LOR-OSEM with an accurate point-spread function (PSF) model. Two forms of LROC analysis were performed. First, a channelized nonprewhitened (CNPW) observer was used to optimize processing parameters (number of iterations, post-reconstruction filter) for the human observer study. Human observers then rated each image and selected the most-likely lesion location. The area under the LROC curve ( A(LROC)) and the probability of correct localization were used as figures-of-merit. The results of the human observer study found no statistically significant difference between FORE and AW-OSEM3D ( A(LROC)=0.41 and 0.36, respectively), an increase in lesion detection performance for LOR-OSEM3D ( A(LROC)=0.45, p=0.076), and additional improvement with the use of the PSF model ( A(LROC)=0.55, p=0.024). The numerical CNPW observer provided the same rankings among algorithms, but obtained different values of A(LROC). These results show improved lesion detection performance for the reconstruction algorithms with more sophisticated statistical and imaging models as compared to the previous-generation algorithms.

Fast, accurate and shift-varying line projections for iterative reconstruction using the GPU

List-mode processing provides an efficient way to deal with sparse projections in iterative image reconstruction for emission tomography. An issue often reported is the tremendous amount of computation required by such algorithm. Each recorded event requires several back- and forward line projections. We investigated the use of the programmable graphics processing unit (GPU) to accelerate the line-projection operations and implement fully-3D list-mode ordered-subsets expectation-maximization for positron emission tomography (PET). We designed a reconstruction approach that incorporates resolution kernels, which model the spatially-varying physical processes associated with photon emission, transport and detection. Our development is particularly suitable for applications where the projection data is sparse, such as high-resolution, dynamic, and time-of-flight PET reconstruction. The GPU approach runs more than 50 times faster than an equivalent CPU implementation while image quality and accuracy are virtually identical. This paper describes in details how the GPU can be used to accelerate the line projection operations, even when the lines-of-response have arbitrary endpoint locations and shift-varying resolution kernels are used. A quantitative evaluation is included to validate the correctness of this new approach.

Exact and approximate Fourier rebinning of PET data from time-of-flight to non time-of-flight

The image reconstruction problem for fully 3D TOF PET is challenging because of the large data sizes involved. One approach to this problem is to first rebin the data into one of the following lower dimensional formats: 2D TOF, 3D non TOF or 2D non TOF. Here we present a unified framework based on a generalized projection slice theorem for TOF data that can be used to compute each of these mappings. We use this framework to develop approaches for rebinning into non TOF formats without significant loss of information. We first derive the exact mappings and then describe approximations which address the missing data problem for oblique sinograms. We evaluate the performance of approximate rebinning using Monte Carlo simulations. Our results show that rebinning into non TOF sinograms retains significant SNR advantages over sinograms collected without TOF information.

Practical method for radioactivity distribution analysis in small-animal PET cancer studies

We present a practical method for radioactivity distribution analysis in small-animal tumors and organs using positron emission tomography imaging with a calibrated source of known activity and size in the field of view. We reconstruct the imaged mouse together with a source under the same conditions, using an iterative method, Maximum likelihood expectation-maximization with system modeling, capable of delivering high-resolution images. Corrections for the ratios of geometrical efficiencies, radioisotope decay in time and photon attenuation are included in the algorithm. We demonstrate reconstruction results for the amount of radioactivity within the scanned mouse in a sample study of osteolytic and osteoblastic bone metastasis from prostate cancer xenografts. Data acquisition was performed on the small-animal PET system, which was tested with different radioactive sources, phantoms and animals to achieve high sensitivity and spatial resolution. Our method uses high-resolution images to determine the volume of organ or tumor and the amount of their radioactivity has the possibility of saving time, effort and the necessity to sacrifice animals. This method has utility for prognosis and quantitative analysis in small-animal cancer studies, and will enhance the assessment of characteristics of tumor growth, identifying metastases, and potentially determining the effectiveness of cancer treatment. The possible application for this technique could be useful for the organ radioactivity dosimetry studies.

Rotate-and-slant projector for fast LOR-based fully-3-D iterative PET reconstruction

One of the greatest challenges facing iterative fully-3-D positron emission tomography (PET) reconstruction is the issue of long reconstruction times due to the large number of measurements for 3-D mode as compared to 2-D mode. A rotate-and-slant projector has been developed that takes advantage of symmetries in the geometry to compute volumetric projections to multiple oblique sinograms in a computationally efficient manner. It is based upon the 2-D rotation-based projector using the three-pass method of shears, and it conserves the 2-D rotator computations for multiple projections to each oblique sinogram set. The projector is equally applicable to both conventional evenly-spaced projections and unevenly-spaced line-of-response (LOR) data. The LOR-based version models the location and orientation of the individual LORs (i.e., the arc-correction), providing an ordinary Poisson reconstruction framework. The projector was implemented in C with several optimizations for speed, exploiting the vertical symmetry of the oblique projection process, depth compression, and array indexing schemes which maximize serial memory access. The new projector was evaluated and compared to ray-driven and distance-driven projectors using both analytical and experimental phantoms, and fully-3-D iterative reconstructions with each projector were also compared to Fourier rebinning with 2-D iterative reconstruction. In terms of spatial resolution, contrast, and background noise measures, 3-D LOR-based iterative reconstruction with the rotate-and-slant projector performed as well as or better than the other methods. Total processing times, measured on a single cpu Linux workstation, were approximately 10x faster for the rotate-and-slant projector than for the other 3-D projectors studied. The new projector provided four iterations fully-3-D ordered-subsets reconstruction in as little as 15 s--approximately the same time as FORE + 2-D reconstruction. We conclude that the rotate-and-slant projector is a viable option for fully-3-D PET, offering quality statistical reconstruction in times only marginally slower than 2-D or rebinning methods.

Positron emission tomography/computed tomography

Accurate anatomical localization of functional abnormalities obtained with the use of positron emission tomography (PET) is known to be problematic. Although tracers such as (18)F-fluorodeoxyglucose ((18)F-FDG) visualize certain normal anatomical structures, the spatial resolution is generally inadequate for accurate anatomic localization of pathology. Combining PET with a high-resolution anatomical imaging modality such as computed tomography (CT) can resolve the localization issue as long as the images from the two modalities are accurately coregistered. However, software-based registration techniques have difficulty accounting for differences in patient positioning and involuntary movement of internal organs, often necessitating labor-intensive nonlinear mapping that may not converge to a satisfactory result. Acquiring both CT and PET images in the same scanner obviates the need for software registration and routinely provides accurately aligned images of anatomy and function in a single scan. A CT scanner positioned in line with a PET scanner and with a common patient couch and operating console has provided a practical solution to anatomical and functional image registration. Axial translation of the couch between the 2 modalities enables both CT and PET data to be acquired during a single imaging session. In addition, the CT images can be used to generate essentially noiseless attenuation correction factors for the PET emission data. By minimizing patient movement between the CT and PET scans and accounting for the axial separation of the two modalities, accurately registered anatomical and functional images can be obtained. Since the introduction of the first PET/CT prototype more than 6 years ago, numerous patients with cancer have been scanned on commercial PET/CT devices worldwide. The commercial designs feature multidetector spiral CT and high-performance PET components. Experience has demonstrated an increased level of accuracy and confidence in the interpretation of the combined study as compared with studies acquired separately, particularly in distinguishing pathology from normal, physiologic tracer uptake and precisely localizing abnormal foci. Combined PET/CT scanners represent an important evolution in technology that has helped to bring molecular imaging to the forefront in cancer diagnosis, staging and therapy monitoring.

Statistical image reconstruction from correlated data with applications to PET

Most statistical reconstruction methods for emission tomography are designed for data modeled as conditionally independent Poisson variates. In reality, due to scanner detectors, electronics and data processing, correlations are introduced into the data resulting in dependent variates. In general, these correlations are ignored because they are difficult to measure and lead to computationally challenging statistical reconstruction algorithms. This work addresses the second concern, seeking to simplify the reconstruction of correlated data and provide a more precise image estimate than the conventional independent methods. In general, correlated variates have a large non-diagonal covariance matrix that is computationally challenging to use as a weighting term in a reconstruction algorithm. This work proposes two methods to simplify the use of a non-diagonal covariance matrix as the weighting term by (a) limiting the number of dimensions in which the correlations are modeled and (b) adopting flexible, yet computationally tractable, models for correlation structure. We apply and test these methods with simple simulated PET data and data processed with the Fourier rebinning algorithm which include the one-dimensional correlations in the axial direction and the two-dimensional correlations in the transaxial directions. The methods are incorporated into a penalized weighted least-squares 2D reconstruction and compared with a conventional maximum a posteriori approach.

Objective and subjective comparison of standard 2-D and fully 3-D reconstructed data on a PET/CT system

OBJECTIVE

The relative advantage of fully 3-D versus 2-D mode for whole-body imaging is currently the focus of considerable expert debate. The nature of 3-D PET acquisition for FDG PET/CT theoretically allows a shorter scan time and improved efficiency of FDG use than in the standard 2-D acquisition. We therefore objectively and subjectively compared standard 2-D and fully 3-D reconstructed data for FDG PET/CT on a research PET/CT system.

MATERIALS AND METHODS

In a total of 36 patients (mean 58.9 years, range 17.3-78.9 years; 21 male, 15 female) referred for known or suspected malignancy, FDG PET/CT was performed using a research PET/CT system with advanced detector technology with improved sensitivity and spatial resolution. After 45 min uptake, a low-dose CT (40 mAs) from head to thigh was performed followed by 2-D PET (emission 3 min per field) and 3-D PET (emission 1.5 min per field) with both seven slices overlap to cover the identical anatomical region. Acquisition time was therefore 50% less (seven fields; 21 min vs. 10.5 min). PET data was acquired in a randomized fashion, so in 50% of the cases 2-D data was acquired first. CT data was used for attenuation correction. 2-D (OSEM) and 3-D PET images were iteratively reconstructed. Subjective analysis of 2-D and 3-D images was performed by two readers in a blinded, randomized fashion evaluating the following criteria: sharpness of organs (liver, chest wall/lung), overall image quality and detectability and dignity of each identified lesion. Objective analysis of PET data was investigated measuring maximum standard uptake value with lean body mass (SUV(max,LBM)) of identified lesions.

RESULTS

On average, per patient, the SUV(max) was 7.86 (SD 7.79) for 2-D and 6.96 (SD 5.19) for 3-D. On a lesion basis, the average SUV(max) was 7.65 (SD 7.79) for 2-D and 6.75 (SD 5.89) for 3-D. The absolute difference on a paired t-test of SUV 3-D-2-D based on each measured lesion was significant with an average of -0.956 (P=0.002) and an average of -0.884 on a patient base (P<0.05). With 3-D the SUV(max) decreased by an average of 5.2% for each lesion, and an average of 6.0% for each patient. Subjective analysis showed fair inter-observer agreement regarding detectability (kappa=0.24 for 3-D; 0.36 for 3-D) and dignity (kappa=0.44 for 3-D and 0.4 for 2-D) of the lesions. There was no significant diagnostic difference between 3-D and 2-D. Only in one patient, a satellite liver metastasis of a colon cancer was missed in 3-D and detected only in 2-D. On average, the overall image quality for 3-D images was equal (in 24%) or inferior (in 76%) compared to 2-D.

CONCLUSION

A possible major advantage of 3-D data acquisition is the faster patient throughput with a 50% reduction in scan time. The fully 3-D reconstruction technique has overcome the technical drawbacks of current 3-D imaging technique. In our limited number of patients there was no significant diagnostic difference between 2-D and fully 3-D.

Analytical calculation of volumes-of-intersection for iterative, fully 3-D PET reconstruction

Use of iterative algorithms to reconstruct three-dimensional (3-D) positron emission tomography (PET) data requires the computation of the system probability matrix. The pure geometrical contribution can easily be approximated by the length-of-intersection (LOI) between lines-of-response (LOR) and individual voxels. However, more accurate geometrical projectors are desirable. Therefore, we have developed a fast method for the analytical calculation of the 3-D shape and volume of volumes-of-intersection (VOI). This method provides an alternative robust projector with a uniformly continuous sampling of the image space. The enhanced calculation effort is facilitated by using several speedup techniques. Exploiting intrinsic symmetry relations and the sparseness of the system matrix allows to create an efficiently compressed matrix which can be precomputed and completely stored in memory. In addition, a new voxel addressing scheme has been implemented. This scheme avoids time-consuming symmetry transformations of voxel addresses by using an octant-wise symmetrically ordered field of voxels. The above methods have been applied for a fully 3-D, iterative reconstruction of 3-D sinograms recorded with a Siemens/CTI ECAT HR+ PET scanner. A comparison of the performance of the reconstruction using LOI weighting and VOI weighting is presented.

Modeling and incorporation of system response functions in 3-D whole body PET

Appropriate application of spatially variant system models can correct for degraded resolution response and mispositioning errors. This paper explores the detector blurring component of the system model for a whole body positron emission tomography (PET) system and extends this factor into a more general system response function to account for other system effects including the influence of Fourier rebinning (FORE). We model the system response function as a three-dimensional (3-D) function that blurs in the radial and axial dimension and is spatially variant in radial location. This function is derived from Monte Carlo simulations and incorporates inter-crystal scatter, crystal penetration, and the blurring due to the FORE algorithm. The improved system model is applied in a modified ordered subsets expectation maximization (OSEM) algorithm to reconstruct images from rebinned, fully 3-D PET data. The proposed method effectively removes the spatial variance in the resolution response, as shown in simulations of point sources. Furthermore, simulation and measured studies show the proposed method improves quantitative accuracy with a reduction in tumor bias compared to conventional OSEM on the order of 10%-30% depending on tumor size and smoothing parameter.

Analytical reconstruction of deconvolved Fourier rebinned PET sinograms

Fully 3D PET data are often rebinned into 2D data sets in order to avoid computationally intensive fully 3D reconstruction. Then, conventional 2D reconstruction techniques are employed to obtain images from the rebinned data. In a common scenario, 2D filtered back projection (FBP) is applied to Fourier rebinned (FORE) data. This approach is suboptimal because FBP is based on an idealized mathematical model of the data and cannot account for the statistical structure of data and noise. FORE data contain some blur in all three dimensions in comparison to conventional 2D PET data. In this work, we propose methods for approximating this blur in the sinogram domain due to FORE through its point spread function (PSF). We also explore simple methods for deconvolving the rebinned data with this PSF to restore it to a more ideal state prior to FBP. Our results show that deconvolution of the approximate transaxial PSF yields no improvement. When low image noise levels are required for detection tasks, the deconvolution of the axial PSF does not provide adequate resolution or quantitative benefits to justify its application. When accurate quantitation is required and higher noise levels are acceptable, the deconvolution of the axial PSF leads to considerable gains (30%) in accuracy over conventional FORE+FBP at matched noise levels.

Small animal positron emission tomography in food sciences

Positron emission tomography (PET) is a 3-dimensional imaging technique that has undergone tremendous developments during the last decade. Non-invasive tracing of molecular pathways in vivo is the key capability of PET. It has become an important tool in the diagnosis of human diseases as well as in biomedical and pharmaceutical research. In contrast to other imaging modalities, radiotracer concentrations can be determined quantitatively. By application of appropriate tracer kinetic models, the rate constants of numerous different biological processes can be determined. Rapid progress in PET radiochemistry has significantly increased the number of biologically important molecules labelled with PET nuclides to target a broader range of physiologic, metabolic, and molecular pathways. Progress in PET physics and technology strongly contributed to better scanners and image processing. In this context, dedicated high resolution scanners for dynamic PET studies in small laboratory animals are now available. These developments represent the driving force for the expansion of PET methodology into new areas of life sciences including food sciences. Small animal PET has a high potential to depict physiologic processes like absorption, distribution, metabolism, elimination and interactions of biologically significant substances, including nutrients, 'nutriceuticals', functional food ingredients, and foodborne toxicants. Based on present data, potential applications of small animal PET in food sciences are discussed.

Catabolism of native and oxidized low density lipoproteins: in vivo insights from small animal positron emission tomography studies

The human organism is exposed to numerous processes that generate reactive oxygen species (ROS). ROS may directly or indirectly cause oxidative modification and damage of proteins. Protein oxidation is regarded as a crucial event in the pathogenesis of various diseases ranging from rheumatoid arthritis to Alzheimer's disease and atherosclerosis. As a representative example, oxidation of low density lipoprotein (LDL) is regarded as a crucial event in atherogenesis. Data concerning the role of circulating oxidized LDL (oxLDL) in the development and outcome of diseases are scarce. One reason for this is the shortage of methods for direct assessment of the metabolic fate of circulating oxLDL in vivo. We present an improved methodology based on the radiolabelling of apoB-100 of native LDL (nLDL) and oxLDL, respectively, with the positron emitter fluorine-18 ((18)F) by conjugation with N-succinimidyl-4-[(18)F]fluorobenzoate ([(18)F]SFB). Radiolabelling of both nLDL and oxLDL using [(18)F]SFB causes neither additional oxidative structural modifications of LDL lipids and proteins nor alteration of their biological activity and functionality, respectively, in vitro. The method was further evaluated with respect to the radiopharmacological properties of both [(18)F]fluorobenzoylated nLDL and oxLDL by biodistribution studies in male Wistar rats. The metabolic fate of [(18)F]fluorobenzoylated nLDL and oxLDL in rats in vivo was further delineated by dynamic positron emission tomography (PET) using a dedicated small animal tomograph (spatial resolution of 2 mm). From this study we conclude that the use of [(18)F]FB-labelled LDL particles is an attractive alternative to, e.g., LDL iodination methods, and is of value to characterize and to discriminate the kinetics and the metabolic fate of nLDL and oxLDL in small animals in vivo.

Fourier rebinning of time-of-flight PET data

This paper explores fast reconstruction strategies for 3D time-of-flight (TOF) positron emission tomography (PET), based on 2D data rebinning. Starting from pre-corrected 3D TOF data, a rebinning algorithm estimates for each transaxial slice the 2D TOF sinogram that would have been acquired by a single-ring scanner. The rebinned sinograms can then be reconstructed using any algorithm for 2D TOF reconstruction. We introduce TOF-FORE, an approximate rebinning algorithm obtained by extending the Fourier rebinning method for non-TOF data. In addition, we identify two partial differential equations that must be satisfied by consistent 3D TOF data, and use them to derive exact rebinning algorithms and to characterize the degree of the approximation in TOF-FORE. Numerical simulations demonstrate that TOF-FORE is more accurate than two different TOF extensions of the single-slice rebinning method, and suggest that TOF-FORE will be a valuable tool for practical TOF PET in the range of axial apertures and time resolutions typical of current scanners.