Monte Carlo-based Reconstruction Research Articles

Monte Carlo-based scatter correction for the SMARTZOOM collimator

BackgroundMyocardial perfusion imaging is a commonly performed SPECT protocol and hence it would be beneficial if its scan duration could be shortened. For traditional gamma cameras, two developments have separately shown to allow for a shortened scan duration: (i) reconstructing with Monte Carlo-based scatter correction instead of dual-energy window scatter correction and (ii) acquiring projections with the SMARTZOOM collimator instead of a parallel-hole collimator. This study investigates which reduction in scan duration can be achieved when both methods are combined in a single system.ResultsThe SMARTZOOM collimator was implemented in a Monte Carlo-based reconstruction package and the implementation was validated through image quality phantom experiments. The potential for scan duration reduction was evaluated with a phantom configuration that is realistic for myocardial perfusion imaging. The original reconstruction quality was achieved in 76 ± 8% of the original scan duration when switching from dual-energy window scatter correction to Monte Carlo-based scatter correction. The original reconstruction quality was achieved in 56 ± 13% of the original scan duration when switching from the parallel-hole to the SMARTZOOM collimator. After combining both methods in a single system, the original reconstruction quality was achieved in 34 ± 7% of the original scan duration.ConclusionsMonte Carlo-based scatter correction combined with the SMARTZOOM collimator can further decrease the scan duration in myocardial perfusion imaging.

Establishment of a clinical SPECT/CT protocol for imaging of 161Tb

BackgroundIt has been proposed, and preclinically demonstrated, that 161Tb is a better alternative to 177Lu for the treatment of small prostate cancer lesions due to its high emission of low-energy electrons. 161Tb also emits photons suitable for single-photon emission computed tomography (SPECT) imaging. This study aims to establish a SPECT protocol for 161Tb imaging in the clinic.Materials and methodsOptimal settings using various γ-camera collimators and energy windows were explored by imaging a Jaszczak phantom, including hollow-sphere inserts, filled with 161Tb. The collimators examined were extended low-energy general purpose (ELEGP), medium-energy general purpose (MEGP), and low-energy high resolution (LEHR), respectively. In addition, three ordered subset expectation maximization (OSEM) algorithms were investigated: attenuation-corrected OSEM (A-OSEM); attenuation and dual- or triple-energy window scatter-corrected OSEM (AS-OSEM); and attenuation, scatter, and collimator-detector response-corrected OSEM (ASC-OSEM), where the latter utilized Monte Carlo-based reconstruction. Uniformity corrections, using intrinsic and extrinsic correction maps, were also investigated. Image quality was assessed by estimated recovery coefficients (RC), noise, and signal-to-noise ratio (SNR). Sensitivity was determined using a circular flat phantom.ResultsThe best RC and SNR were obtained at an energy window between 67.1 and 82.1 keV. Ring artifacts, caused by non-uniformity, were removed with extrinsic uniformity correction for the energy window between 67.1 and 82.1 keV, but not with intrinsic correction. Analyzing the lower energy window between 48.9 and 62.9 keV, the ring artifacts remained after uniformity corrections. The recovery was similar for the different collimators when using a specific OSEM reconstruction. Recovery and SNR were highest for ASC-OSEM, followed by AS-OSEM and A-OSEM. When using the optimized parameter setting, the resolution of 161Tb was higher than for 177Lu (8.4 ± 0.7 vs. 10.4 ± 0.6 mm, respectively). The sensitivities for 161Tb and 177Lu were 7.41 and 8.46 cps/MBq, respectively.ConclusionSPECT with high resolution is feasible with 161Tb; however, extrinsic uniformity correction is recommended to avoid ring artifacts. The LEHR collimator was the best choice of the three tested to obtain a high-resolution image. Due to the complex emission spectrum of low-energy photons, window-based scatter correction had a minor impact on the image quality compared to using attenuation correction only. On the other hand, performing attenuation, scatter, and collimator-detector correction clearly improved image quality. Based on these data, SPECT-based dosimetry for 161Tb-labeled radiopharmaceuticals is feasible.

Simultaneous 166Ho/99mTc dual-isotope SPECT with Monte Carlo-based downscatter correction for automatic liver dosimetry in radioembolization

BackgroundIntrahepatic dosimetry is paramount to optimize radioembolization treatment accuracy using radioactive holmium-166 microspheres (166Ho). This requires a practical protocol that combines quantitative imaging of microsphere distribution with automated and robust delineation of the volumes of interest. To this end, we propose a dual isotope single photon emission computed tomography (SPECT) protocol based on 166Ho therapeutic microspheres and technetium-99 m (99mTc) stannous phytate, which accumulates in healthy liver tissue. This protocol may allow accurate and automatic estimation of tumor-absorbed dose and healthy liver-absorbed dose. The current study focuses on a Monte Carlo-based reconstruction framework that inherently corrects for scatter crosstalk between the 166Ho and 99mTc imaging. To demonstrate the feasibility of the method, it is evaluated with realistic phantom experiments and patient data.MethodsThe Utrecht Monte Carlo System (UMCS) was extended to include detailed modeling of crosstalk interactions between 99mTc and 166Ho. First, 99mTc images were reconstructed including energy window-based corrections for 166Ho downscatter. Next, 99mTc downscatter in the 81-keV 166Ho window was Monte Carlo simulated to allow quantitative reconstruction of the 166Ho images. The accuracy of the 99mTc-downscatter modeling was evaluated by comparing measurements with simulations. In addition, the ratio between 99mTc and 166Ho yielding the best 166Ho dose estimates was established and the quantitative accuracy was reported.ResultsGiven the same level of activity, 99mTc contributes twice as many counts to the 81-keV window than 166Ho, and four times as many counts to the 140-keV window, applying a 166Ho/99mTc ratio of 5:1 yielded a high accuracy in both 166Ho and 99mTc reconstruction. Phantom experiments revealed that the accuracy of quantitative 166Ho activity recovery was reduced by 10% due to the presence of 99mTc. Twenty iterations (8 subsets) of the SPECT/CT reconstructions were considered feasible for clinical practice. Applicability of the proposed protocol was shown in a proof-of-concept case.ConclusionA novel 166Ho/99mTc dual-isotope protocol for automatic dosimetry compensates accurately for downscatter and allows for the addition of 99mTc without compromising 166Ho SPECT image quality.

Accelerated SPECT image reconstruction with FBP and an image enhancement convolutional neural network

BackgroundMonte Carlo-based iterative reconstruction to correct for photon scatter and collimator effects has been proven to be superior over analytical correction schemes in single-photon emission computed tomography (SPECT/CT), but it is currently not commonly used in daily clinical practice due to the long associated reconstruction times. We propose to use a convolutional neural network (CNN) to upgrade fast filtered back projection (FBP) image quality so that reconstructions comparable in quality to the Monte Carlo-based reconstruction can be obtained within seconds.ResultsA total of 128 technetium-99m macroaggregated albumin pre-treatment SPECT/CT scans used to guide hepatic radioembolization were available. Four reconstruction methods were compared: FBP, clinical reconstruction, Monte Carlo-based reconstruction, and the neural network approach. The CNN generated reconstructions in 5 sec, whereas clinical reconstruction took 5 min and the Monte Carlo-based reconstruction took 19 min. The mean squared error of the neural network approach in the validation set was between that of the Monte Carlo-based and clinical reconstruction, and the lung shunting fraction difference was lower than 2 percent point. A phantom experiment showed that quantitative measures required in radioembolization were accurately retrieved from the CNN-generated reconstructions.ConclusionsFBP with an image enhancement neural network provides SPECT reconstructions with quality close to that obtained with Monte Carlo-based reconstruction within seconds.

Fast quantitative reconstruction with focusing collimators for liver SPECT

BackgroundGeneration of a SPECT scan during procedure may aid in the optimization of treatments as liver radioembolization by offering image-guided dosimetry. This, however, requires both shortened acquisition times and fast quantitative reconstruction. Focusing collimators increase sensitivity and thus may speed up imaging. Monte Carlo-based iterative reconstruction has shown to provide quantitative results for parallel hole collimators but may be slow. The purpose of this work is to develop fast Monte Carlo-based reconstruction for focusing collimators and to evaluate the impact of reconstruction and collimator choice on quantitative accuracy of liver dosimetry by means of simulations.ResultsThe developed fast Monte Carlo simulator was found to accurately generate projections compared to a full Monte Carlo simulation, providing projections in several seconds instead of several days. Monte Carlo-based scatter correction was superior to other scatter correction methods in describing recovered activity and reached similar noise levels as dual-energy window scatter correction. Although truncation artifacts were present in the cone beam collimator (50 cm), the region inside the field of view (FOV) could be reconstructed without loss of accuracy. Provided the object to image is inside the FOV, the focusing collimator with 50 cm focal distance could retrieve the same noise levels as a parallel hole collimator in 68% of the total scanning time, the multifocal collimator in 73% of the time, and the 100-cm focal distance collimator in 84% of the time.ConclusionFocusing collimators combined with Monte Carlo-based reconstruction have the ability to enable quantitative imaging of the FOV in a significantly shorter timeframe. The proposed approach to the forward projector will additionally make it possible to reconstruct within minutes. These are crucial steps in moving toward real-time dosimetry during interventions.

Monte Carlo-based quantitative pinhole SPECT reconstruction using a ray-tracing back-projector

BackgroundMonte Carlo simulations provide accurate models of nuclear medicine imaging systems as they can properly account for the full physics of photon transport. The accuracy of the model included in the maximum-likelihood–expectation-maximization (ML-EM) reconstruction limits the overall accuracy of the reconstruction results. In this paper, we present a Monte Carlo-based ML-EM reconstruction method for pinhole single-photon emission computed tomography (SPECT) that has been incorporated into the SIMIND Monte Carlo program. The Monte Carlo-based model, which accounts for all of the physical and geometrical characteristics of the camera system, is used in the forward-projection step of the reconstruction, while a simpler model based on ray-tracing is used for back-projection. The aim of this work was to investigate the quantitative accuracy of this combination of forward- and back-projectors in the clinical pinhole camera GE Discovery NM 530c.ResultsThe total activity was estimated in 99mTc-filled spheres with volumes between 0.5 and 16 mL. The total sphere activity was generally overestimated but remained within 10% of the reference activity defined by the phantom preparation. The recovered activity converged towards the reference activity as the number of iterations increased. Furthermore, the recovery of the activity concentrations within the physical boundaries of the spheres increased with increasing sphere volume. Additionally, the Monte Carlo-based reconstruction enabled recovery of the true activity concentration in the myocardium of a cardiac phantom mounted in a torso phantom regardless of whether the torso was empty or water-filled. A qualitative comparison to data reconstructed using the clinical reconstruction algorithm showed that the two methods performed similarly, although the images reconstructed using the clinical software were more uniform due to the incorporation of noise regularization and post-filtration in that reconstruction technique.ConclusionsWe developed a Monte Carlo-based reconstruction method for pinhole SPECT and evaluated it using phantom measurements. The combination of a Monte Carlo-based forward-projector and a simplified analytical ray-tracing back-projector produced quantitative images of acceptable image quality. No explicit calibration is necessary in this method since the forward-projector model maintains a relationship between the number of counts and activity.

Characterization of the ringing artifacts in rotator‐based reconstruction with Monte Carlo‐based resolution compensation for PET

Studies have shown that Monte Carlo-based reconstruction could effectively improve the image quality of positron emission tomography. The authors have previously used a Gaussian rotator-based algorithm to efficiently reduce the computational cost for system matrix (SM) calculation and to meet the large memory requirements for SM storage. However, pronounced ringing artifacts were observed in the reconstructed image. In this article, the authors investigated and characterized these artifacts. The authors proposed an "ideal" rotator and used it as a baseline in the artifacts evaluation. This ideal rotator produces perfectly rotated images. The Gaussian rotator method was evaluated by a full system model and a partial system model without positron range and acolinearity, which could be compensated for by the blurring of the Gaussian rotator for 18F studies. Noiseless data, Monte Carlo simulation data, as well as acquired experimental data were used to quantitatively characterize the behavior of the artifacts. The study of the noiseless data indicated that the artifacts were mainly attributed to the rotator, which further blurred the simulated system responses. The simulation study suggested that the artifacts become less pronounced and not quantitatively significant in practice. This result is consistent with the experimental data study. Better contrast recovery was achieved with an over-compensated system model. Traditionally, an undercompensated system model was preferred to avoid artifacts. The authors' studies suggest that the Gaussian rotator with the full system model yields the best image quality among the evaluated methods with considerably reduced quantitative error and quantitatively insignificant artifacts in practice. The authors' investigation indicated that a moderately overcompensated system model (about 2 mm FWHM in this study) yielded better contrast recovery and quantitatively insignificant artifacts in practice.

Coronal Heat: Solar Loop Temperatures from TRACE Triple-Filter Data

The Transition Region and Coronal Explorer (TRACE) has state-of-the-art spatial resolution and shows the most detailed images of coronal loops ever observed. The temperatures of these loops are primarily derived from the 171 to 195 Å filter ratio, with data from the third filter at 284 Å used by several authors to improve the precision of the derived temperatures. Most of these studies assume that the plasma is isothermal and model the loops primarily as uniform temperature structures with footpoint-dominated heating. However, these triple-filter data are insufficient to constrain the plasma temperature and cannot be used to determine the isothermality or otherwise of coronal loop structures. We show this explicitly by constructing differential emission measures with these same triple-filter data using a sophisticated Markov-chain Monte Carlo-based reconstruction algorithm. We find that these TRACE data cannot, in general, limit the temperature distribution for coronal loop plasma. In other words, many different temperature distributions (isothermal, broad, sloped, etc.) can reproduce the observed fluxes, and the TRACE coronal data alone cannot determine which of these distributions represents the actual coronal plasma.