S. Bullich

Evaluation of the geometric, scatter and septal penetration components in fan beam collimators using Monte Carlo simulation

The quantitative analysis of SPECT data requires an accurate determination of the collimator point spread function (PSF). The aim of this work is to characterize the PSFs of fan beam and parallel collimators by using Monte Carlo simulation. Given a particular collimator configuration, a detailed hexagonal hole array is generated and information describing its geometry is stored in a look-up table. When a photon crosses the collimator front plane, a forty-hole array is placed around its impact position using this table. Each photon is then tracked up to the detector surface by using the Monte Carlo code PENELOPE and its associated geometry handling routines. Particle counters are defined that score the probability of impact on the detector as a function of the final photon position. Four sets of counters are employed so as to differentiate contributions to the geometric, septal penetration, coherent (Rayleigh) and incoherent (Compton) scatter components. Furthermore, sensitivity quantification and pulse-height energy spectra are calculated for different source locations. Monte Carlo results have been compared with sensitivity values obtained experimentally and good agreement was found. The authors results show that for /sup 99m/Tc imaging, the geometric component represents about 95% of the fan beam PSF, whereas the incoherent scattering component is negligible.

Biodistribution and Radiation Dosimetry of the Glycine Transporter-1 Ligand 11C-GSK931145 Determined from Primate and Human Whole-Body PET

Purpose11C-GSK931145 is a novel radioligand suitable for imaging the glycine transporter 1 (GlyT-1) in brain. In the present study, human dosimetry is estimated from baboon and human biodistribution data.ProceduresThree baboons and eight healthy human volunteers underwent whole-body positron emission tomography (PET) scans. Human dosimetry was estimated using three different region-of-interest (ROI) delineation methods that ranged in their complexity and execution time: ROIs drawn on anterior-posterior compressed PET images, on subsamples of the organs, and covering the whole-organ. Residence times for each organ were calculated as the area under the time-activity curves divided by the injected activity. Radiation dose estimates were calculated from organ residence times using the OLINDA/EXM software package.ResultsThe overall distribution of activity was similar in baboons and humans. Early scans presented high activity in the liver, and moderate activity in the lungs and kidneys. The principal route of clearance was intestinal and no urinary excretion was observed. The limiting organ with the highest radiation-absorbed dose was the liver. The mean effective dose in humans was 4.02xa0μSv/MBq (male phantom) and 4.95xa0μSv/MBq (female phantom) (ROIs drawn on subsamples of the organs). The human effective dose estimated from baboon data was ~15% larger than the effective dose estimated from human data.ConclusionHuman PET imaging of the glycine transporter-1 with 11C-GSK931145 results in a moderate effective human radiation dose, which allows for multiple PET examinations in the same individual. Among the three methods compared to delineate ROIs, the organ subsampling method shows the best balance between quantitative accuracy and practical application.

Modelling of high-energy contamination in SPECT imaging using Monte Carlo simulation

123I is a commonly used radioisotope employed in neurotransmitter SPECT studies. In addition to an intense line at 159 keV, the decay scheme of this radioisotope includes a low yield (~3%) of higher energy photons which have a non-negligible contribution to the final image when low-energy high-resolution (LEHR) collimators are used. This contribution of high-energy photons may achieve ~28% of the total counts in the projections. The aim of this work is to model each energy component of the high-energy Point Spread Function (hPSF) for fan-beam LEHR collimators in order to develop faster Monte Carlo (MC) simulations of high-energy ray contamination. The modelling of hPSF was based on the results of simulating photons through the collimator-detector system using the MC code PENELOPE. Since low-energy PSFs models for fan-beam collimators must tend to a Gaussian distribution, we use the same function for the hPSF modelling for high-energy photons. The parameters of these Gaussian functions were obtained by minimizing the root mean square (RMS) error between each simulated hPSF and the function g(x,y) using the efficiency of the simulated hPSFs as a constraint The RMS attained with fit of g(x,y) to the simulated hPSFs was always smaller than ~2% of the mean efficiency per pixel of the image. A very strong dependence of the efficiency on the type and thickness of the backscatter material behind the crystal was found. The hPSFs were parameterized for a wide range of energies, ranging from 350 keV to 538 keV. Our results indicate that Gaussian distributions approximate in a suitable way the hPSF responses for fan-beam collimators. This model will be an important tool to accelerate MC simulations of radiolabelled compounds which emit medium- or high-energy rays

Impact of scatter correction on D2 receptor occupancy measurements using 123I-IBZM SPECT: comparison to 11C-Raclopride PET.

Reported values of D(2) receptor occupancy (RO) achieved by antipsychotic drugs tend to be lower when measured with (123)I-IBZM SPECT than with (11)C-Raclopride PET. Image degrading factors such as attenuation, distance-dependent collimator response and scatter could account for this difference. While attenuation correction is routinely applied to SPECT images, the other degradations are not usually accounted for. The aim of this work was to assess the impact of scatter correction on D(2) RO quantification with (123)I-IBZM SPECT, and to compare the results of both corrected and un-corrected SPECT values with (11)C-Raclopride PET measurements. Phantom experiments as well as within-subject human data from a previous study were used for this purpose. SPECT images were reconstructed using filtered back-projection including attenuation correction (FBP(A)), ordered subsets expectation maximization including attenuation and point spread function corrections (OSEM(A+PSF)) and ordered subsets expectation maximization including attenuation, point spread function and scatter corrections (OSEM(A+PSF+SCT)). PET images were reconstructed using the FBP algorithm and corrected for attenuation, scatter, random coincidences and dead time. Quantification of receptor availability was performed using the tissue ratio at pseudoequilibrium for SPECT, and the simplified reference tissue model (SRTM) for PET. Analysis was performed using both occipital cortex (occ) and cerebellum (cer) as reference regions for both modalities. When images were reconstructed using FBP(A), SPECT D(2) RO values were significantly lower as compared with PET leading to a D(2) RO difference of -20% (CI(95%): -13, -27%) (occ) and -23% (CI(95%): -14, -31%) (cer). When images were reconstructed using OSEM(A+PSF), SPECT D(2) RO values were also lower as compared with PET leading to a D(2) RO difference of -21% (CI(95%): -14, -27%) (occ) and -24% (CI(95%): -18, -30%) (cer). When images were reconstructed using OSEM(A+PSF+SCT), the D(2) RO bias was reduced to -6% (CI(95%): 0, -13%) (occ) and -11% (CI(95%): -4, -18%) (cer). These data suggest that the scatter correction plays a major role in explaining the differences between D(2) RO measurements using (123)I-IBZM SPECT and (11)C-Raclopride PET.

Modeling of high-energy contamination in SPECT imaging using Monte Carlo simulation

/sup 123/I is a commonly used radioisotope employed in neurotransmitter SPECT studies. In addition to an intense line at 159 keV, the decay scheme of this radioisotope includes a low yield (/spl sim/3%) of higher energy photons which make a non-negligible contribution to the final image when low-energy high-resolution (LEHR) collimators are used. This contribution of high-energy photons may reach /spl sim/28% of the total counts in the projections. The aim of this work is to model each energy component of the high-energy Point Spread Function (hPSF) for fan-beam LEHR collimators in order to develop fast Monte Carlo (MC) simulations of high-energy ray contamination. The modeling of hPSF was based on the results of simulating photons through the collimator-detector system using the MC code PENELOPE. Since low-energy PSFs models for fan-beam collimators tend to a Gaussian distribution, we use the same function for the hPSF modeling for high-energy photons. The parameters of these Gaussian functions (g(x,y)) were obtained by minimizing the root mean square error (RMS) using the sensitivity of the simulated hPSFs as a constraint. The hPSFs were parameterized for a range of energies between 350 keV and 538 keV. The RMS attained after fitting of g(x,y) to the simulated hPSFs was always smaller than /spl sim/2% of the mean sensitivity per pixel of the image. A strong dependence of the sensitivity on the type and thickness of the backscatter material behind the crystal was found. Our results indicate that Gaussian distributions approximate the hPSF responses for fan-beam collimators. This model will be an important tool to accelerate MC simulations of radiolabeled compounds which emit medium- or high-energy rays.

Neurotransmission SPECT and MR registration combining mutual and gradient information.

Purpose: Image registration is important in functional image analysis. In neurotransmission single photon emission tomography (nSPECT), specific uptake sites can be accurately localized by superimposing the SPECT study onto a high-resolution structural image such as a magnetic resonance(MR) of the subject. Mutual-information (MI)-based algorithms are usually employed for this purpose. Nevertheless, nSPECT/MR registration using MI is often limited by the low count rates present in nSPECT. Several works have proposed extensions of the MI measures to include gradient information (GI) from the images but their performance has not been evaluated in SPECT studies. Methods: In this work, the accuracy of the MI including gradient information (MIG) was compared with the standard MI using data from healthy volunteers and data simulating a specific uptake reduction using three different radioligands: I 123 -IBZM , I 123 -ADAM , I 123 -R 91150 . Results: The results showed that MIG-based registration yielded better accuracy than MI. The MIG-based similarity measures were less sensitive to sparse sampling and diminished computational time without a substantial decrease in registration accuracy. Conclusions: Accuracy of nSPECT/MR registration is improved when gradient information is included in the MI-based algorithm, which makes MIG-based registration potentially useful for clinical applications.

Impact of Region-of-Interest Delineation Methods, Reconstruction Algorithms, and Intra- and Inter-Operator Variability on Internal Dosimetry Estimates Using PET

PurposeHuman dosimetry studies play a central role in radioligand development for positron emission tomography (PET). Drawing regions of interest (ROIs) on the PET images is used to measure the dose in each organ. In the study aspects related to ROI delineation methods were evaluated for two radioligands of different biodistribution (intestinal vs urinary).ProceduresPET images were simulated from a human voxel-based phantom. Several ROI delineation methods were tested: antero-posterior projections (AP), 3D sub-samples of the organs (S), and a 3D volume covering the whole-organ (W). Inter- and intra-operator variability ROI drawing was evaluated by using human data.ResultsThe effective dose estimates using S and W methods were comparable to the true values. AP methods overestimated (49xa0%) the dose for the radioligand with intestinal biodistribution. Moreover, the AP method showed the highest inter-operator variability: 11xa0±xa01xa0%.ConclusionsThe sub-sampled organ method showed the best balance between quantitative accuracy and inter- and intra-operator variability.