Design study of an in situ PET scanner for use in proton beam therapy.

Proton beam therapy can deliver a high radiation dose to a tumor without significant damage to surrounding healthy tissue or organs. One way of verifying the delivered dose distribution is to image the short-lived positron emitters produced by the proton beam as it travels through the patient. A potential solution to the limitations of PET imaging in proton beam therapy is the development of a high sensitivity, in situ PET scanner that starts PET imaging almost immediately after patient irradiation while the patient is still lying on the treatment bed. A partial ring PET design is needed for this application in order to avoid interference between the PET detectors and the proton beam, as well as restrictions on patient positioning on the couch. A partial ring also allows us to optimize the detector separation (and hence the sensitivity) for different patient sizes. Our goal in this investigation is to evaluate an in situ PET scanner design for use in proton therapy that provides tomographic imaging in a partial ring scanner design using time-of-flight (TOF) information and an iterative reconstruction algorithm. GEANT4 simulation of an incident proton beam was used to produce a positron emitter distribution, which was parameterized and then used as the source distribution inside a water-filled cylinder for EGS4 simulations of a PET system. Design optimization studies were performed as a function of crystal type and size, system timing resolution, scanner angular coverage and number of positron emitter decays. Data analysis was performed to measure the accuracy of the reconstructed positron emitter distribution as well as the range of the positron emitter distribution. We simulated scanners with varying crystal sizes (2-4 mm) and type (LYSO and LaBr(3)) and our results indicate that 4 mm wide LYSO or LaBr(3) crystals (resulting in 4-5 mm spatial resolution) are adequate; for a full-ring, non-TOF scanner we predict a low bias (<0.6 mm) and a good precision (<1 mm) in the estimated range relative to the simulated positron distribution. We then varied the angular acceptance of the scanner ranging from 1/2 to 2/3 of 2π; a partial ring TOF imaging with good timing resolution (≤600 ps) is necessary to produce accurate tomographic images. A two-third ring scanner with 300 ps timing resolution leads to a bias of 1.0 mm and a precision of 1.4 mm in the range estimate. With a timing resolution of 600 ps, the bias increases to 2.0 mm while the precision in the range estimate is similar. For a half-ring scanner design, more distortions are present in the image, which is characterized by the increased error in the profile difference estimate. We varied the number of positron decays imaged by the PET scanner by an order of magnitude and we observe some decrease in the precision of the range estimate for lower number of decays, but all partial ring scanner designs studied have a precision ≤1.5 mm. The largest number tested, 150 M total positron decays, is considered realistic for a clinical fraction of delivered dose, while the range of positron decays investigated in this work covers a variable number of situations corresponding to delays in scan start time and the total scan time. Thus, we conclude that for partial ring systems, an angular acceptance of at least 1/2 (of 2π) together with timing resolution of 300 ps is needed to achieve accurate and precise range estimates. With 600 ps timing resolution an angular acceptance of 2/3 (of 2π) is required to achieve satisfactory range estimates. These results indicate that it would be feasible to develop a partial-ring dedicated PET scanner based on either LaBr(3) or LYSO to accurately characterize the proton dose for therapy planning.

The imaging performance of a LaBr3-based PET scanner.

A prototype time-of-flight (TOF) PET scanner based on cerium-doped lanthanum bromide [LaBr(3) (5% Ce)] has been developed. LaBr(3) has a high light output, excellent energy resolution and fast timing properties that have been predicted to lead to good image quality. Intrinsic performance measurements of spatial resolution, sensitivity and scatter fraction demonstrate good conventional PET performance; the results agree with previous simulation studies. Phantom measurements show the excellent image quality achievable with the prototype system. Phantom measurements and corresponding simulations show a faster and more uniform convergence rate, as well as more uniform quantification, for TOF reconstruction of the data, which have 375 ps intrinsic timing resolution, compared to non-TOF images. Measurements and simulations of a hot and cold sphere phantom show that the 7% energy resolution helps to mitigate residual errors in the scatter estimate because a high energy threshold (>480 keV) can be used to restrict the amount of scatter accepted without a loss of true events. Preliminary results with incorporation of a model of detector blurring in the iterative reconstruction algorithm not only show improved contrast recovery but also point out the importance of an accurate resolution model of the tails of LaBr(3)'s point spread function. The LaBr(3) TOF-PET scanner demonstrated the impact of superior timing and energy resolutions on image quality.

Performance of a whole-body PET scanner using curve-plate NaI(Tl) detectors.

UNLABELLED: A whole-body PET scanner, without interplane septa, has been designed to achieve high performance in clinical applications. The C-PET scanner, an advancement of the PENN PET scanners, is unique in the use of 6 curved NaI(Tl) detectors (2.54 cm thick). The scanner has a ring diameter of 90 cm, a patient port diameter of 56 cm, and an axial field of view of 25.6 cm. A (137)Cs point source is used for transmission scans. METHODS: Following the protocols of the International Electrotechnical Commission ([IEC] 61675-1) and the National Electrical Manufacturers Association ([NEMA] NU-2-1994 and an updated version, NU2-2001), point and line sources, as well as uniform cylinders, were used to determine the performance characteristics of the C-PET scanner. An image-quality phantom and patient data were used to evaluate image quality under clinical scanning conditions. Data were rebinned with Fourier rebinning into 2-dimensional (slice-oriented) datasets and reconstructed with an iterative reconstruction algorithm. RESULTS: The spatial resolution for a point source in the transaxial direction was 4.6 mm (full width at half maximum) at the center, and the axial resolution was 5.7 mm. For the NU2-1994 analysis, the sensitivity was 12.7 cps/Bq/mL (444 kcps/microCi/mL), the scatter fraction was 25%, and the peak noise equivalent count rate (NEC) for a uniform cylinder (diameter = 20 cm, length = 19 cm) was 49 kcps at an activity concentration of 11.2 kBq/mL. For the IEC protocol, the peak NEC was 41 kcps at 12.3 kBq/mL, and for the NU2-2001 protocol, the peak NEC was 14 kcps at 3.8 kBq/mL. The NU2-2001 NEC value differed significantly because of differences in the data analysis and the use of a 70-cm-long phantom. CONCLUSION: Compared with previous PENN PET scanners, the C-PET, with its curved detectors and improvements in pulse shaping, integration dead time, and triggering, has an improved count-rate capability and spatial resolution. With the refinements in the singles transmission technique and iterative reconstruction, image quality is improved and scan time is shortened. With single-event transmission scans interleaved between sequential emission scans, a whole-body study can be completed in <1 h. Overall, C-PET is a cost-effective PET scanner that performs well in a broad variety of clinical applications.

Device-dependent activity estimation and decay correction of radionuclide mixtures with application to Tc-94m PET studies.

Multi-instrument activity estimation and decay correction techniques were developed for radionuclide mixtures, motivated by the desire for accurate quantitation of Tc-94m positron emission tomography (PET) studies. Tc-94m and byproduct Tc isotopes were produced by proton irradiation of enriched Mo-94 and natural Mo targets. Mixture activities at the end of bombardment were determined with a calibrated high purity germanium detector. The activity fractions of the greatest mixture impurities relative to 100% for Tc-94m averaged 10.0% (Tc-94g) and 3.3% (Tc-93) for enriched targets and 10.1% (Tc-94g), 11.0% (Tc-95), 255.8% (Tc-96m), and 7.2% (Tc-99m) for natural targets. These radioisotopes have different half-lives (e.g., 52.5 min for Tc-94m, 293 min for Tc-94g), positron branching ratios (e.g., 0.72 for Tc-94m, 0.11 for Tc-94g) and gamma ray emissions for themselves and their short-lived, excited Mo daughters. This complicates estimation of injected activity with a dose calibrator, in vivo activity with PET and blood sample activity with a gamma counter. Decay correction using only the Tc-94m half-life overestimates activity and is inadequate. For this reason analytic formulas for activity estimation and decay correction of radionuclide mixtures were developed. Isotope-dependent sensitivity factors for a PET scanner, dose calibrator, and gamma counter were determined using theoretical sensitivity models and fits of experimental decay curves to sums of exponentials with fixed decay rates. For up to 8 h after the end of bombardment with activity from enriched and natural Mo targets, decay-corrected activities were within 3% of the mean for three PET studies of a uniform cylinder, within 3% of the mean for six dose calibrator decay studies, and within 6% of the mean for four gamma counter decay studies. Activity estimation and decay correction for Tc-94m mixtures enable routine use of Tc-94m in quantitative PET, as illustrated by application to a canine Tc-94m sestamibi study.

Performance characteristics of the 3-D OSEM algorithm in the reconstruction of small animal PET images. Ordered-subsets expectation-maximixation.

Rat brain images acquired with a small animal positron emission tomography (PET) camera and reconstructed with the three-dimensional (3-D) ordered-subsets expectation-maximization (OSEM) algorithm with resolution recovery have better quality when the brain is imaged by itself than when inside the head with surrounding background activity. The purpose of this study was to characterize the dependence of this effect on the level of background activity, attenuation, and scatter. Monte Carlo simulations of the imaging system were performed. The coefficient of variation from replicate images, full-width at half-maximum (FWHM) from point sources and image profile fitting, and image contrast and uniformity were used to evaluate algorithm performance. A rat head with the typical levels of five and ten times the brain activity in the surrounding background requires additional iterations to achieve the same resolution as the brain-only case at a cost of 24% and 64% additional noise, respectively. For the same phantoms, object scatter reduced contrast by 3%-5%. However, attenuation degraded resolution by 0.2 mm and was responsible for up to 12% nonuniformity in the brain images suggesting that attenuation correction is useful. Given the effects of emission and attenuation distribution on both resolution and noise, simulations or phantom studies should be used for each imaging situation to select the appropriate number of OSEM iterations to achieve the desired resolution-noise levels.

Noise characteristics of 3-D and 2-D PET images.

We analyzed the noise characteristics of two-dimensional (2-D) and three-dimensional (3-D) images obtained from the GE Advance positron emission tomography (PET) scanner. Three phantoms were used: a uniform 20-cm phantom, a 3-D Hoffman brain phantom, and a chest phantom with heart and lung inserts. Using gated acquisition, we acquired 20 statistically equivalent scans of each phantom in 2-D and 3-D modes at several activity levels. From these data, we calculated pixel normalized standard deviations (NSD's), scaled to phantom mean, across the replicate scans, which allowed us to characterize the radial and axial distributions of pixel noise. We also performed sequential measurements of the phantoms in 2-D and 3-D modes to measure noise (from interpixel standard deviations) as a function of activity. To compensate for the difference in axial slice width between 2-D and 3-D images (due to the septa and reconstruction effects), we developed a smoothing kernel to apply to the 2-D data. After matching the resolution, the ratio of image-derived NSD values (NSD2D/NSD3D)2 averaged throughout the uniform phantom was in good agreement with the noise equivalent count (NEC) ratio (NEC3D/NEC2D). By comparing different phantoms, we showed that the attenuation and emission distributions influence the spatial noise distribution. The estimates of pixel noise for 2-D and 3-D images produced here can be applied in the weighting of PET kinetic data and may be useful in the design of optimal dose and scanning requirements for PET studies. The accuracy of these phantom-based noise formulas should be validated for any given imaging situation, particularly in 3-D, if there is significant activity outside the scanner field of view.

Axial slice width in 3D PET: characterization and potential improvement with axial interleaving.

The axial slice width for the GE Advance PET scanner has previously been reported to be worse for 3D acquisitions than for 2D. The goals of this study were to investigate the source(s) of this observed difference and to assess whether the 3D axial slice width could be significantly improved by acquisition and simultaneous reconstruction of axially interleaved data. The axial slice width was measured for the three acquisition modes of the Advance scanner ('standard' high-sensitivity 2D, high-resolution 2D and 3D), with the septa both extended and retracted. A significant degradation in the axial slice width for 3D compared with that for high-sensitivity 2D mode was seen. Near the centre, this difference can largely be attributed to septa collimation effects of the 2D data. At larger radial positions, axial mispositioning of cross-coincidences in 2D acquisitions overshadows the effects of septa collimation, while 3D reconstruction effects also become more important. The axial slice width was estimated to improve by 0.3-1.1 mm with interleaving. This modest improvement would be accompanied by an increase in image noise, since an axial filter with a higher cut-off would be required in the 3D reconstruction to achieve this resolution in the image.

Optimization of noninvasive activation studies with 15O-water and three-dimensional positron emission tomography.

We investigated the effects of varying the injected dose, speed of injection, and scan duration to maximize the sensitivity of noninvasive activation studies with 15O-water and three-dimensional positron emission tomography. A covert word generation task was used in four subjects with bolus injections of 2.5 to 3D mCi of 15O-water. The noise equivalent counts (NEC) for the whole brain peaked at an injected dose of 12 to 15 mCi. This was lower than expected from phantom studies, presumably because of the effect of radioactivity outside of the brain. A 10 mCi injection gave an NEC of 92.4 +/- 2.2% of the peak value. As the scan duration increased from 60 to 90 to 120 seconds, the areas of activation decreased in size or were no longer detected. Therefore, we selected a 1 minute scan using 10 mCi for bolus injections. We then performed simulation studies to evaluate, for a given CBF change, the effect on signal-to-noise ratio (S/N) of longer scan duration with slow tracer infusions. Using a measured arterial input function from a bolus injection, new input functions for longer duration injections and the corresponding tissue data were simulated. Combining information about image noise derived from Hoffman brain phantom studies with the simulated tissue data allowed calculation of the S/N for a given CBF change. The simulation shows that a slow infusion permits longer scan acquisitions with only a small loss in S/N. This allows the investigator to choose the injection duration, and thus the time period during which scan values are sensitive to regional CBF.

A head motion measurement system suitable for emission computed tomography.

Subject motion during brain imaging studies can adversely affect the images through loss of resolution and other artifacts related to movement. We have developed and tested a device to measure head motion externally in real-time during emission computed tomographic (ECT) brain imaging studies, to be used eventually to correct ECT data for that motion. The system is based on optical triangulation of three miniature lights affixed to the patient's head and viewed by two position-sensitive detectors. The computer-controlled device converts the three sets of lamp positions into rotational and translational coordinates every 0.7 seconds. When compared against a mechanical test fixture, the optical system was found to be linear and accurate with minimal crosstalk between the coordinates. In a study of two subjects, comparing the angular motions measured by the optical device and a commercially available electromagnetic motion detector, the two systems agreed well, with an root mean square (rms) difference of less than 0.6 degree for all rotations.

Precision and accuracy of regional radioactivity quantitation using the maximum likelihood EM reconstruction algorithm.

The imaging characteristics of maximum likelihood (ML) reconstruction using the EM algorithm for emission tomography have been extensively evaluated. There has been less study of the precision and accuracy of ML estimates of regional radioactivity concentration. The authors developed a realistic brain slice simulation by segmenting a normal subject's MRI scan into gray matter, white matter, and CSF and produced PET sinogram data with a model that included detector resolution and efficiencies, attenuation, scatter, and randoms. Noisy realizations at different count levels were created, and ML and filtered backprojection (FBP) reconstructions were performed. The bias and variability of ROI values were determined. In addition, the effects of ML pixel size, image smoothing and region size reduction were assessed. Hit estimates at 3,000 iterations (0.6 sec per iteration on a parallel computer) for 1-cm(2) gray matter ROIs showed negative biases of 6%+/-2% which can be reduced to 0%+/-3% by removing the outer 1-mm rim of each ROI. FBP applied to the full-size ROIs had 15%+/-4% negative bias with 50% less noise than hit. Shrinking the FBP regions provided partial bias compensation with noise increases to levels similar to ML. Smoothing of ML images produced biases comparable to FBP with slightly less noise. Because of its heavy computational requirements, the ML algorithm will be most useful for applications in which achieving minimum bias is important.

An approximation formula for the variance of PET region-of-interest values.

An approximation formula for the variance of positron emission tomography (PET) region-of-interest (ROI) values has been developed, implemented, and evaluated. This formula does not require access to the original projection data and is therefore convenient for routine use. The formula was derived by applying successive approximations to the filtered-backprojection reconstruction algorithm. ROI variance is estimated from the product of mean pixel variance within the region and a term accounting for the intercorrelation of all pixel pairs inside the region. The formula accounts for radioactivity distribution, attenuation, randoms, scatter, deadtime, detector normalization, scan length, decay, and reconstruction filter. The algorithm was tested by comparison to the exact ROI variance as calculated with Huesman's algorithm. Tests with scan data from phantoms, animals, and humans obtained on the Scanditronix PC2048-15B tomograph showed the approximation formula to be accurate to within +/-10%

Performance standards in positron emission tomography.

A standard set of performance measurements is proposed for use with positron emission tomographs. This set of measurements has been developed jointly by the Computer and Instrumentation Council of the Society of Nuclear Medicine and the National Electrical Manufacturers Association. The measurements include tests of spatial resolution, scatter fraction, sensitivity, count rate losses and randoms, uniformity, scatter correction, attenuation correction, and count rate linearity correction.

Factors affecting accuracy and precision in PET volume imaging.

Volume imaging positron emission tomographic (PET) scanners with no septa and a large axial acceptance angle offer several advantages over multiring PET scanners. A volume imaging scanner combines high sensitivity with fine axial sampling and spatial resolution. The fine axial sampling minimizes the partial volume effect, which affects the measured concentration of an object. Even if the size of an object is large compared to the slice spacing in a multiring scanner, significant variation in the concentration is measured as a function of the axial position of the object. With a volume imaging scanner, it is necessary to use a three-dimensional reconstruction algorithm in order to avoid variations in the axial resolution as a function of the distance from the center of the scanner. In addition, good energy resolution is needed in order to use a high energy threshold to reduce the coincident scattered radiation.

Unified deadtime correction model for PET.

A model of deadtime for emission and transmission scans in positron emission tomography (PET) scanners with two-dimensional detectors has been developed. The model takes into account coincidence losses due to singles losses and multiple events, as well as mispositioning errors at higher count rates caused by pulse pile-up, within a detector block. The model is applicable to emission distributions and to spatially varying singles distributions seen with a rotating pin transmission source. An automatic procedure to determine the parameters of this model based on decaying emission studies has also been developed. Different singles dead time factors are required for emission and blank distributions due to differences in their energy spectra. The model was tested on emission and pin transmission data taken on the Scanditronix PC2048-15B scanner.

Continuous-slice PENN-PET: a positron tomograph with volume imaging capability.

The PENN-PET scanner consists of six hexagonally arranged position-sensitive Nal(TI) detectors. This design offers high spatial resolution in all three dimensions, high sampling density along all three axes without scanner motion, a large axial acceptance angle, good energy resolution, and good timing resolution. This results in three-dimensional imaging capability with high sensitivity and low scatter and random backgrounds. The spatial resolution is 5.5 mm (FWHM) in all directions near the center. The true sensitivity, for a brain-sized object, is a maximum of 85 kcps/microCi/ml and the scatter fraction is a minimum of 10%, both depending on the lower level energy threshold. The scanner can handle up to 5 mCi in the field of view, at which point the randoms equal the true coincidences and the detectors reach their count rate limit. We have so far acquired [18F]FDG brain studies and cardiac studies, which show the applicability of our scanner for both brain and whole-body imaging. With the results to date, we feel that this design results in a simple yet high performance scanner which is applicable to many types of static and dynamic clinical studies.

Quantitative comparison of cerebral glucose metabolic rates from two positron emission tomographs.

The rapid progress in positron emission tomography technology has created the dilemma of how to compare data from old and new tomographs. We examined cerebral metabolic data from two scanners, with different spatial resolutions and methods of attenuation correction, to see if data from the lower resolution tomograph (ECAT II) could be "corrected" and then compared to data from the higher resolution scanner (Scanditronix PC1024-7B). Nine subjects were scanned on both tomographs after a single injection of [18F]2-fluoro-2-deoxy-D-glucose. Regional and lobar gray matter metabolic rates for glucose were obtained from comparable images from each scanner. Ratios of lobar to global gray matter metabolism also were calculated. Regression coefficients and percent differences were computed to compare ECAT II and PC1024 data. Twenty-four of the 36 regions showed significant regression slopes, and PC1024 measures of glucose utilization ranged from 30% to 120% higher than those from the ECAT II. Lobar differences between the two machines were less variable (50% to 80%), and ratios generally differed by only +/- 5%. Since there was no simple and consistent relation between regional metabolic rates on the two tomographs, an overall adjustment of regional ECAT values for comparison to PC1024 values would be impossible. A region-by-region adjustment would be necessary. On the other hand, ratios are sufficiently similar that direct comparisons could be made.

A method for postinjection PET transmission measurements with a rotating source.

A method is presented for obtaining accurate positron emission tomography transmission measurements after tracer injection. A transmission scan is performed using a rotating source immediately before or after a conventional emission scan. Sinogram windowing, which removes most scattered and random coincidences, also removes most of the emission counts contaminating the transmission measurement. Data from the emission scan can be used to subtract the remaining emission counts to produce accurate transmission measurements. For studies with moderate to low emission count rates (e.g., fluorodeoxyglucose) there is little increase in noise in the resulting attenuation correction factors. This method was tested in experiments with phantoms and a rotating source simulator and validated against conventional ring transmission measurements. Applications of the technique can significantly shorten the time between transmission and emission studies, and thereby reduce the likelihood of patient motion and increase scanning throughput.

The effect of energy window on cardiac ejection fraction.

ECG gated gamma-ray energy spectra from the left ventricle were created each 50 msec during the cardiac cycle. Nine of ten subjects were studied with a nonimaging Nal probe, and the tenth with a high-resolution Germanium detector. Placing multiple energy windows over the energy spectra, EF was found to vary with the energy window selected. Moving a 20% window across the photopeak produced a roughly linear increase in EF with energy (2.3 EF units per 10 keV increase in energy) in eight of the ten subjects. Dividing the photopeak into a low (126-140 keV) and high-energy (140-154 keV) portion gave significantly different EFs (high energy exceeding low energy by 17%). Increasing the width of a narrow window centered about the photopeak produced negligible change in EF. Examining the energy spectra showed that the small-angle scattered radiation (126-139 keV) was proportionately greater at end systole than at end diastole, after normalizing the spectra to the same photopeak area.

Treatment of axial data in three-dimensional PET.

Improved axial spatial resolution in positron emission tomography (PET) scanners will lead to reduced sensitivity unless the axial acceptance angle for the coincidences is kept constant. A large acceptance angle, however, violates assumptions made in most reconstruction algorithms, which reconstruct parallel independent slices, rather than a three-dimensional volume. Two methods of treating the axial information from a volume PET scanner are presented. Qualitative and quantitative errors introduced by the approximations are examined for simulated objects with sharp boundaries and for a more anatomically realistic distribution with smooth activity gradients.

A silicone gel phantom suitable for multimodality imaging.

A silicone gel phantom material is described. It can be molded into any shape and it can contain internal cavities. The material is selfsealing allowing the cavities to be filled with liquids. A simple geometric phantom consisting of a rectangular solid with a single internal spherical cavity was fabricated and imaged using MRI, CT, and PET. Properties important for imaging are described.