Jonathan A. Disselhorst

Spatial Resolution and Sensitivity of the Inveon Small-Animal PET Scanner

The Inveon small-animal PET scanner is characterized by a large, 127-mm axial length and a 161-mm crystal ring diameter. The associated high sensitivity is obtained by using all lines of response (LORs) up to the maximum ring difference (MRD) of 79, for which the most oblique LORs form acceptance angles of 38.3° with transaxial planes. The result is 2 phenomena that are normally not encountered in PET scanners: a parallax or depth-of-interaction effect in the axial direction and the breakdown of Fourier rebinning (FORE). Both effects cause a deterioration of axial spatial resolution. Limiting the MRD to smaller values reduces this axial blurring at the cost of sensitivity. Alternatively, 3-dimensional (3D) reconstruction techniques can be used in which the rebinning step is absent. The aim of this study was to experimentally determine the spatial resolution and sensitivity of the Inveon for its whole field of view (FOV). Methods: Spatial resolution and sensitivity were measured using filtered backprojection (FBP) with FORE, FBP with LOR angle-weighted adapted FORE (AFORE), and 3D ordered-subset expectation maximization followed by maximum a posteriori reconstruction (OSEM3D/MAP). Results: Tangential and radial full width at half maximum (FWHM) showed almost no dependence on the MRD using FORE and FBP. Tangential FWHMs were 1.5 mm in the center of the FOV (CFOV) and 1.8 mm at the edge of the FOV (EFOV). Radial FWHMs were 1.5 and 3.0 mm in the CFOV and EFOV, respectively. In contrast, axial FWHMs increased with the MRD and ranged between 1.1 and 2.0 mm in the CFOV and between 1.5 and 2.7 mm in the EFOV for a MRD between 1 and 79. AFORE improved the axial resolution for a large part of the FOV, but image noise increased. OSEM3D/MAP yielded uniform spatial resolution in all directions, with an average FWHM of 1.65 ± 0.06 mm. Sensitivity in the CFOV for the default energy and coincidence time window was 0.068; peak sensitivity was 0.111. Conclusion: The Inveon showed high spatial resolution and high sensitivity, both of which can be maintained using OSEM3D/MAP reconstruction instead of rebinning and 2D algorithms.

Image-quality assessment for several positron emitters using the NEMA NU 4-2008 standards in the Siemens Inveon small-animal PET scanner.

The positron emitters 18F, 68Ga, 124I, and 89Zr are all relevant in small-animal PET. Each of these radionuclides has different positron energies and ranges and a different fraction of single photons emitted. Average positron ranges larger than the intrinsic spatial resolution of the scanner (for 124I and 68Ga) will deteriorate the effective spatial resolution and activity recovery coefficient (RC) for small lesions or phantom structures. The presence of single photons (for 124I and 89Zr) could increase image noise and spillover ratios (SORs). Methods: Image noise, expressed as percentage SD in a uniform region (%SD), RC, and SOR (in air and water) were determined using the NEMA NU 4 small-animal image-quality phantom filled with 3.7 MBq of total activity of 18F, 68Ga, 124I, or 89Zr. Filtered backprojection (FBP), ordered-subset expectation maximization in 2 dimensions, and maximum a posteriori (MAP) reconstructions were compared. In addition to the NEMA NU 4 image-quality parameters, spatial resolutions were determined using small glass capillaries filled with these radionuclides in a water environment. Results: The %SD for 18F, 68Ga, 124I, and 89Zr using FBP was 6.27, 6.40, 6.74, and 5.83, respectively. The respective RCs were 0.21, 0.11, 0.12, and 0.19 for the 1-mm-diameter rod and 0.97, 0.65, 0.64, and 0.88 for the 5-mm-diameter rod. SORs in air were 0.01, 0.03, 0.04, and 0.01, respectively, and in water 0.02, 0.10, 0.13, and 0.02. Other reconstruction algorithms gave similar differences between the radionuclides. MAP produced the highest RCs. For the glass capillaries using FBP, the full widths at half maximum for 18F, 68Ga, 124I, and 89Zr were 1.81, 2.46, 2.38, and 1.99 mm, respectively. The corresponding full widths at tenth maximum were 3.57, 6.52, 5.87, and 4.01 mm. Conclusion: With the intrinsic spatial resolution (≈1.5 mm) of this latest-generation small-animal PET scanner, the finite positron range has become the limiting factor for the overall spatial resolution and activity recovery in small structures imaged with 124I and 68Ga. The presence of single photons had only a limited effect on the image noise. MAP, as compared with the other reconstruction algorithms, increased RC and decreased %SD and SOR.

Pretargeted Immuno–Positron Emission Tomography Imaging of Carcinoembryonic Antigen–Expressing Tumors with a Bispecific Antibody and a 68Ga- and 18F-Labeled Hapten Peptide in Mice with Human Tumor Xenografts

18F-Fluorodeoxyglucose (18F-FDG) is the most common molecular imaging agent in oncology, with a high sensitivity and specificity for detecting several cancers. Antibodies could enhance specificity; therefore, procedures were developed for radiolabeling a small (∼1451 Da) hapten peptide with 68Ga or 18F to compare their specificity with 18F-FDG for detecting tumors using a pretargeting procedure. Mice were implanted with carcinoembryonic antigen (CEA; CEACAM5)–expressing LS174T human colonic tumors and a CEA-negative tumor, or an inflammation was induced in thigh muscle. A bispecific monoclonal anti-CEA × anti-hapten antibody was given to mice, and 16 hours later, 5 MBq of 68Ga- or 18F-labeled hapten peptides were administered intravenously. Within 1 hour, tissues showed high and specific targeting of 68Ga-IMP-288, with 10.7 ± 3.6% ID/g uptake in the tumor and very low uptake in normal tissues (e.g., tumor-to-blood ratio of 69.9 ± 32.3), in a CEA-negative tumor (0.35 ± 0.35% ID/g), and inflamed muscle (0.72 ± 0.20% ID/g). 18F-FDG localized efficiently in the tumor (7.42 ± 0.20% ID/g) but also in the inflamed muscle (4.07 ± 1.13% ID/g) and in several normal tissues; thus, pretargeted 68Ga-IMP-288 provided better specificity and sensitivity. Positron emission tomography (PET)/computed tomography images reinforced the improved specificity of the pretargeting method. 18F-labeled IMP-449 distributed similarly in the tumor and normal tissues as the 68Ga-labeled IMP-288, indicating that either radiolabeled hapten peptide could be used. Thus, pretargeted immuno-PET does exceptionally well with short-lived radionuclides and is a highly sensitive procedure that is more specific than 18F-FDG-PET. Mol Cancer Ther; 9(4); 1019–27. ©2010 AACR.

Principles of PET/MR Imaging

Hybrid PET/MR systems have rapidly progressed from the prototype stage to systems that are increasingly being used in the clinics. This review provides an overview of developments in hybrid PET/MR systems and summarizes the current state of the art in PET/MR instrumentation, correction techniques, and data analysis. The strong magnetic field requires considerable changes in the manner by which PET images are acquired and has led, among others, to the development of new PET detectors, such as silicon photomultipliers. During more than a decade of active PET/MR development, several system designs have been described. The technical background of combined PET/MR systems is explained and related challenges are discussed. The necessity for PET attenuation correction required new methods based on MR data. Therefore, an overview of recent developments in this field is provided. Furthermore, MR-based motion correction techniques for PET are discussed, as integrated PET/MR systems provide a platform for measuring motion with high temporal resolution without additional instrumentation. The MR component in PET/MR systems can provide functional information about disease processes or brain function alongside anatomic images. Against this background, we point out new opportunities for data analysis in this new field of multimodal molecular imaging.

Using the NEMA NU 4 PET Image Quality Phantom in Multipinhole Small-Animal SPECT

Several commercial small-animal SPECT scanners using multipinhole collimation are presently available. However, generally accepted standards to characterize the performance of these scanners do not exist. Whereas for small-animal PET, the National Electrical Manufacturers Association (NEMA) NU 4 standards have been defined in 2008, such standards are still lacking for small-animal SPECT. In this study, the image quality parameters associated with the NEMA NU 4 image quality phantom were determined for a small-animal multipinhole SPECT scanner. Methods: Multiple whole-body scans of the NEMA NU 4 image quality phantom of 1-h duration were performed in a U-SPECT-II scanner using 99mTc with activities ranging between 8.4 and 78.2 MBq. The collimator contained 75 pinholes of 1.0-mm diameter and had a bore diameter of 98 mm. Image quality parameters were determined as a function of average phantom activity, number of iterations, postreconstruction spatial filter, and scatter correction. In addition, a mouse was injected with 99mTc-hydroxymethylene diphosphonate and was euthanized 6.5 h after injection. Multiple whole-body scans of this mouse of 1-h duration were acquired for activities ranging between 3.29 and 52.7 MBq. Results: An increase in the number of iterations was accompanied by an increase in the recovery coefficients for the small rods (RCrod), an increase in the noise in the uniform phantom region, and a decrease in spillover ratios for the cold-air– and water-filled scatter compartments (SORair and SORwat). Application of spatial filtering reduced image noise but lowered RCrod. Filtering did not influence SORair and SORwat. Scatter correction reduced SORair and SORwat. The effect of total phantom activity was primarily seen in a reduction of image noise with increasing activity. RCrod, SORair, and SORwat were more or less constant as a function of phantom activity. The relation between acquisition and reconstruction settings and image quality was confirmed in the 99mTc-hydroxymethylene diphosphonate mouse scans. Conclusion: Although developed for small-animal PET, the NEMA NU 4 image quality phantom was found to be useful for small-animal SPECT as well, allowing for objective determination of image quality parameters and showing the trade-offs between several of these parameters on variation of acquisition and reconstruction settings.

Comparison of Tumor Uptake Heterogeneity Characterization Between Static and Parametric 18F-FDG PET Images in Non–Small Cell Lung Cancer

18F-FDG PET is well established in the field of oncology for diagnosis and staging purposes and is increasingly being used to assess therapeutic response and prognosis. Many quantitative indices can be used to characterize tumors on 18F-FDG PET images, such as SUVmax, metabolically active tumor volume (MATV), total lesion glycolysis, and, more recently, the proposed intratumor uptake heterogeneity features. Although most PET data considered within this context concern the analysis of activity distribution using images obtained from a single static acquisition, parametric images generated from dynamic acquisitions and reflecting radiotracer kinetics may provide additional information. The purpose of this study was to quantify differences between volumetry, uptake, and heterogeneity features extracted from static and parametric PET images of non–small cell lung carcinoma (NSCLC) in order to provide insight on the potential added value of parametric images. Methods: Dynamic 18F-FDG PET/CT was performed on 20 therapy-naive NSCLC patients for whom primary surgical resection was planned. Both static and parametric PET images were analyzed, with quantitative parameters (MATV, SUVmax, SUVmean, heterogeneity) being extracted from the segmented tumors. Differences were investigated using Spearman rank correlation and Bland–Altman analysis. Results: MATV was slightly smaller on static images (−2% ± 7%), but the difference was not significant (P = 0.14). All derived parameters, including those characterizing tumor functional heterogeneity, correlated strongly between static and parametric images (r = 0.70–0.98, P ≤ 0.0006), exhibiting differences of less than ±25%. Conclusion: In NSCLC primary tumors, parametric and static baseline 18F-FDG PET images provided strongly correlated quantitative features for both standard (MATV, SUVmax, SUVmean) and heterogeneity quantification. Consequently, heterogeneity quantification on parametric images does not seem to provide significant complementary information compared with static SUV images.

Spectral Clustering predicts tumor tissue heterogeneity using dynamic 18F-FDG PET: a complement to the standard compartmental modeling approach

In this study, we described and validated an unsupervised segmentation algorithm for the assessment of tumor heterogeneity using dynamic 18F-FDG PET. The aim of our study was to objectively evaluate the proposed method and make comparisons with compartmental modeling parametric maps and SUV segmentations using simulations of clinically relevant tumor tissue types. Methods: An irreversible 2-tissue-compartmental model was implemented to simulate clinical and preclinical 18F-FDG PET time–activity curves using population-based arterial input functions (80 clinical and 12 preclinical) and the kinetic parameter values of 3 tumor tissue types. The simulated time–activity curves were corrupted with different levels of noise and used to calculate the tissue-type misclassification errors of spectral clustering (SC), parametric maps, and SUV segmentation. The utility of the inverse noise variance– and Laplacian score–derived frame weighting schemes before SC was also investigated. Finally, the SC scheme with the best results was tested on a dynamic 18F-FDG measurement of a mouse bearing subcutaneous colon cancer and validated using histology. Results: In the preclinical setup, the inverse noise variance–weighted SC exhibited the lowest misclassification errors (8.09%–28.53%) at all noise levels in contrast to the Laplacian score–weighted SC (16.12%–31.23%), unweighted SC (25.73%–40.03%), parametric maps (28.02%–61.45%), and SUV (45.49%–45.63%) segmentation. The classification efficacy of both weighted SC schemes in the clinical case was comparable to the unweighted SC. When applied to the dynamic 18F-FDG measurement of colon cancer, the proposed algorithm accurately identified densely vascularized regions from the rest of the tumor. In addition, the segmented regions and clusterwise average time–activity curves showed excellent correlation with the tumor histology. Conclusion: The promising results of SC mark its position as a robust tool for quantification of tumor heterogeneity using dynamic PET studies. Because SC tumor segmentation is based on the intrinsic structure of the underlying data, it can be easily applied to other cancer types as well.

Shortened dynamic (18)F-FDG PET

TO THE EDITOR: With great interest we read a recent article by Strauss et al. (1). The authors describe a support vector machine–based method to predict the parameters of the 2-tissuecompartment model from shortened dynamic 18F-FDG PET acquisitions by analyzing a large database of 1,474 time–activity curves obtained from 539 patients. Shortening the standard 1-h protocol to more convenient acquisition times of less than 30 min would not only improve patient comfort but also reduce demand on camera time and facilitate scheduling of dynamic scans. In this manner, the likelihood that dynamic PET will actually be used for routine imaging purposes would increase. The authors have shown that their method can accurately estimate tumor microparameters using a short dynamic 18F-FDG PET scan. However, we wish to suggest additional analyses. Accumulation of 18F-FDG in a tumor increases with time. Hamberg et al. (2) have shown a continuing rise in standardized uptake value in some lung tumors even several hours after injection. With decreasing blood concentrations, the tumor-to-background ratio continues to increase, but conversely, the decreasing counting rates as a result of the physical decay of 18F dictate an upper limit to the optimal uptake period. Most optimized protocols advise that acquisition of static PET scans begin at least 45 min after administration of 18F-FDG (3,4), and many centers use an uptake period of about 60 min. Volumes of interest (VOIs) to assess uptake or pharmacokinetic parameters are often defined on a threshold basis, such as the 3-dimensional isocontour at 50% of the maximum voxel value within a lesion. Other methods include manually placed VOIs or fixed volumes. These methods have variable advantages and limitations, but all have in common that voxels included in the VOI defined at an earlier time point may differ from those defined in the final time frame. Also, with manually placed VOIs it may be difficult to accurately delineate the lesion, as the contrast is still relatively low at an earlier time point. Consequently, the lesion’s time–activity curve can differ as well, which, in turn, could alter the parameters of the 2-tissue-compartment model. In our experience, the VOI often differs significantly depending on time after injection. The Jaccard index (5) can be used to determine the similarity between 2 VOIs, defined as the number of overlapping voxels divided by the number of voxels in both or any of the VOIs. Comparing VOIs defined in early time frames and the final time frame shows a gradually decreasing similarity. Especially with scans of less than 30 min, the index can become relatively low, because of insufficiently high tumor-to-background ratios. Obviously, with a short dynamic PET acquisition and an additional time frame at 60 min after injection, as also described by Strauss et al., accurate VOI definition is no longer a problem as long as both scans can be registered properly. However, the benefits of a shortened acquisition period would be reduced. Strauss et al. appear to have shortened the dynamic PET scan by removing time points from the original time–activity curves, without redefining the VOIs in the earlier time frames—at least, this is not mentioned in their paper. We would be interested in the combined effect of redefining VOIs on the shortened acquisition and the significantly shorter time–activity curve. When the parameters of the 2-tissue-compartment model can still be estimated with great accuracy, shortened dynamic PET acquisitions could be a valuable addition to standard, static, 18F-FDG PET.

Linking imaging to omics utilizing image-guided tissue extraction

Significance Imaging provides an insight into biological patho-mechanisms of diseases. However, the link between the imaging phenotype and the underlying molecular processes is often not well understood. Methods such as metabolomics and proteomics reveal detailed information about these processes. Unfortunately, they provide no spatial information and thus cannot be easily correlated with functional imaging. We have developed an image-guided milling machine and unique workflows to precisely isolate tissue samples based on imaging data. The tissue samples remain cooled during the entire procedure, preventing sample degradation. This enables us to correlate, at an unprecedented spatial precision, comprehensive imaging information with metabolomics and proteomics data, leading to a better understanding of diseases. Phenotypic heterogeneity is commonly observed in diseased tissue, specifically in tumors. Multimodal imaging technologies can reveal tissue heterogeneity noninvasively in vivo, enabling imaging-based profiling of receptors, metabolism, morphology, or function on a macroscopic scale. In contrast, in vitro multiomics, immunohistochemistry, or histology techniques accurately characterize these heterogeneities in the cellular and subcellular scales in a more comprehensive but ex vivo manner. The complementary in vivo and ex vivo information would provide an enormous potential to better characterize a disease. However, this requires spatially accurate coregistration of these data by image-driven sampling as well as fast sample-preparation methods. Here, a unique image-guided milling machine and workflow for precise extraction of tissue samples from small laboratory animals or excised organs has been developed and evaluated. The samples can be delineated on tomographic images as volumes of interest and can be extracted with a spatial accuracy better than 0.25 mm. The samples remain cooled throughout the procedure to ensure metabolic stability, a precondition for accurate in vitro analysis.

Endothelial Shear Stress: A Critical Determinant of Arterial Remodeling and Arterial Stiffness in Humans. A Carotid 3.0 Tesla MRI Study.

Background—Low endothelial shear stress (ESS) elicits endothelial dysfunction. However, the relationship between ESS and arterial remodeling and arterial stiffness is unknown in humans. We developed a 3.0-T MRI protocol to evaluate the contribution of ESS to arterial remodeling and stiffness. Methods and Results—Fifteen young (aged 26±3 years) and 15 older (aged 57±3 years) healthy volunteers as well as 15 patients with cardiovascular disease (aged 63±10 years) were enrolled. Phase-contrast MRI of the common carotid arteries was used to derive ESS data from the spatial velocity gradients close to the arterial wall. ESS measurements were performed on 3 occasions and showed excellent reproducibility (intraclass correlation coefficient, 0.79). Multiple linear regression analysis accounting for age and blood pressure revealed that ESS was an independent predictor of the following response variables: carotid wall thickness (regression coefficient [b], −0.19 mm2 per N/m2; P=0.02), lumen area (b, −15.5 mm2 per N/m2; P<0.001), and vessel size (b, −24.0 mm2 per N/m2; P<0.001). Segments of the artery wall exposed to lower ESS were significantly thicker than segments exposed to higher ESS within the same artery (P=0.009). Furthermore, ESS was associated with arterial compliance, accounting for age, blood pressure, and wall thickness (b, −0.003 mm2/mm Hg per N/m2; P=0.04). Conclusions—Our carotid MRI data show that ESS is an important determinant of arterial remodeling and arterial stiffness in humans. The data warrant further studies to evaluate use of carotid ESS as a noninvasive tool to improve the understanding of individual cardiovascular disease risk and to assess novel drug therapies in cardiovascular disease prevention.Background— Low endothelial shear stress (ESS) elicits endothelial dysfunction. However, the relationship between ESS and arterial remodeling and arterial stiffness is unknown in humans. We developed a 3.0-T MRI protocol to evaluate the contribution of ESS to arterial remodeling and stiffness. Methods and Results— Fifteen young (aged 26±3 years) and 15 older (aged 57±3 years) healthy volunteers as well as 15 patients with cardiovascular disease (aged 63±10 years) were enrolled. Phase-contrast MRI of the common carotid arteries was used to derive ESS data from the spatial velocity gradients close to the arterial wall. ESS measurements were performed on 3 occasions and showed excellent reproducibility (intraclass correlation coefficient, 0.79). Multiple linear regression analysis accounting for age and blood pressure revealed that ESS was an independent predictor of the following response variables: carotid wall thickness (regression coefficient [b], −0.19 mm2 per N/m2; P =0.02), lumen area (b, −15.5 mm2 per N/m2; P <0.001), and vessel size (b, −24.0 mm2 per N/m2; P <0.001). Segments of the artery wall exposed to lower ESS were significantly thicker than segments exposed to higher ESS within the same artery ( P =0.009). Furthermore, ESS was associated with arterial compliance, accounting for age, blood pressure, and wall thickness (b, −0.003 mm2/mm Hg per N/m2; P =0.04). Conclusions— Our carotid MRI data show that ESS is an important determinant of arterial remodeling and arterial stiffness in humans. The data warrant further studies to evaluate use of carotid ESS as a noninvasive tool to improve the understanding of individual cardiovascular disease risk and to assess novel drug therapies in cardiovascular disease prevention.