3D FBP Research Articles

Information Capacity of Positron Emission Tomography Scanners

Background: The aim of the present study was to assess the upper information content bound of positron emission tomography (PET) images, by means of the information capacity (IC). Methods: The Geant4 Application for the Tomographic Emission (GATE) Monte Carlo (MC) package was used, and reconstructed images were obtained by using the software for tomographic image reconstruction (STIR). The case study for the assessment of the information content was the General Electric (GE) Discovery-ST PET scanner. A thin-film plane source aluminum (Al) foil, coated with a thin layer of silica and with a 18F-fludeoxyglucose (FDG) bath distribution of 1 MBq was used. The influence of the (a) maximum likelihood estimation-ordered subsets-maximum a posteriori probability-one step late (MLE-OS-MAP-OSL) algorithm, using various subsets (1 to 21) and iterations (1 to 20) and (b) different scintillating crystals on PET scanner’s performance, was examined. The study was focused on the noise equivalent quanta (NEQ) and on the single index IC. Images of configurations by using different crystals were obtained after the commonly used 2-dimensional filtered back projection (FBP2D), 3-dimensional filtered back projection re-projection (FPB3DRP) and the (MLE)-OS-MAP-OSL algorithms. Results: Results shown that the images obtained with one subset and various iterations provided maximum NEQ values, however with a steep drop-off after 0.045 cycles/mm. The single index IC data were maximized for the range of 8–20 iterations and three subsets. The PET scanner configuration incorporating lutetium orthoaluminate perovskite (LuAP) crystals provided the highest NEQ values in 2D FBP for spatial frequencies higher than 0.028 cycles/mm. Bismuth germanium oxide (BGO) shows clear dominance against all other examined crystals across the spatial frequency range, in both 3D FBP and OS-MAP-OSL. The particular PET scanner provided optimum IC values using FBP3DRP and BGO crystals (2.4829 bits/mm2). Conclusions: The upper bound of the image information content of PET scanners can be fully characterized and further improved by investigating the imaging chain components through MC methods.

Sub-3 mm, near-200 ps TOF/DOI-PET imaging with monolithic scintillator detectors in a 70 cm diameter tomographic setup

Recently, a monolithic scintillator detector for time-of-flight (TOF)/depth-of-interaction (DOI) positron emission tomography (PET) was developed. It has a detector spatial resolution of ~1.7 mm full-width-at-half-maximum (FWHM), a coincidence resolving time (CRT) of ~215 ps FWHM, and ~4.7 mm FWHM DOI resolution. Here, we demonstrate, for the first time, the imaging performance of this detector in a 70 cm diameter PET geometry. We built a tomographic setup representative of a whole-body clinical scanner, comprising two coaxially rotating arms, each carrying a detector module, and a central, rotating phantom table. The fully automated setup sequentially acquires all possible lines of response (LORs) of a complete detector ring, using a step-and-shoot acquisition approach. The modules contained 2 × 2 detectors, each detector consisting of a 32 mm × 32 mm × 22 mm LYSO crystal and a digital silicon photomultiplier (dSiPM) array. The system spatial resolution was assessed using a Na-22 point source at different radial distances in the field-of-view (FOV). Using 2D filtered back projection (2D FBP, non-TOF), tangential and radial spatial resolutions of ~2.9 mm FWHM were obtained at the center of the FOV. The use of DOI information resulted in almost uniform spatial resolution throughout the FOV up to a radial distance of 25 cm, where the radial and tangential resolution are ~3.3 mm FWHM and ~4.7 mm FWHM, respectively, whereas without DOI the resolution deteriorates to ~9 mm FWHM. Additional measurements were performed with a Na-22 filled Derenzo-like phantom at different locations within the FOV. Images reconstructed with a TOF maximum-likelihood expectation-maximization (TOF ML-EM) algorithm show that the system is able to clearly resolve 3 mm diameter hot rods up to 25 cm radial distance. The excellent and uniform spatial resolution, combined with an energy resolution of 10.2% FWHM and a CRT of ~212 ps FWHM, indicates a great potential for monolithic scintillators as practical high-performance detectors in TOF/DOI-PET systems.

Pre-clinical Positron Emission Tomography Reconstruction Algorithm Effect on Cu-64 ATSM Lesion Hypoxia.

Objective:Application of distinct positron emission tomography (PET) scan reconstruction algorithms can lead to statistically significant differences in measuring lesion functional properties. We looked at the influence of two-dimensional filtered back projection (2D FBP), two-dimensional ordered subset expectation maximization (2D OSEM), three-dimensional ordered subset expectation maximization (3D OSEM) without 3D maximum a posteriori and with (3D OSEM MAP) on lesion hypoxia tracer uptake using a pre-clinical PET scanner.Methods:Reconstructed images of a rodent tumor model bearing P22 carcinosarcoma injected with hypoxia tracer Copper-64-Diacetyl-bis (N4-methylthiosemicarbazone) (i.e. Cu-64 ATSM) were analyzed at 10 minute intervals till 60 minute post injection. Lesion maximum standardized uptake values (SUVmax) and SUVmax/background SUVmean (T/B) were recorded and investigated after application of multiple algorithm and reconstruction parameters to assess their influence on Cu-64 ATSM measurements and associated trends over time.Results:SUVmaxSUVmax or T/B between 2D FBP, exhibited convergence for OSEM reconstructions while ANOVA results showed a significant difference in SUVmax or T/B between 2D FBP, 2D OSEM, 3D OSEM and 3D OSEM MAP reconstructions across all time frames. SUVmax and T/B were greatest in magnitude for 2D OSEM followed by 3D OSEM MAP, 3D OSEM and then 2D FBP at all time frames respectively. Similarly SUVmax and T/B standard deviations (SD) were lowest for 2D OSEM in comparison with other algorithms.Conclusion:Significantly higher magnitude lesion SUVmax and T/B combined with lower SD were observed using 2D OSEM reconstruction in comparison with 2D FBP, 3D OSEM and 3D OSEM MAP algorithms at all time frames. Results are SUVmax or T/B between 2D FBP, consistent with other published studies however more specimens are required for full validation.

Task-based evaluation of a 4D MAP-RBI-EM image reconstruction method for gated myocardial perfusion SPECT using a human observer study

We evaluated the performance of a new 4D image reconstruction method for improved 4D gated myocardial perfusion (MP) SPECT using a task-based human observer study. We used a realistic 4D NURBS-based Cardiac-Torso (NCAT) phantom that models cardiac beating motion. Half of the population was normal; the other half had a regional hypokinetic wall motion abnormality. Noise-free and noisy projection data with 16 gates/cardiac cycle were generated using an analytical projector that included the effects of attenuation, collimator-detector response, and scatter (ADS), and were reconstructed using the 3D FBP without and 3D OS-EM with ADS corrections followed by different cut-off frequencies of a 4D linear post-filter. A 4D iterative maximum a posteriori rescaled-block (MAP-RBI)-EM image reconstruction method with ADS corrections was also used to reconstruct the projection data using various values of the weighting factor for its prior. The trade-offs between bias and noise were represented by the normalized mean squared error (NMSE) and averaged normalized standard deviation (NSDav), respectively. They were used to select reasonable ranges of the reconstructed images for use in a human observer study. The observers were trained with the simulated cine images and were instructed to rate their confidence on the absence or presence of a motion defect on a continuous scale. We then applied receiver operating characteristic (ROC) analysis and used the area under the ROC curve (AUC) index. The results showed that significant differences in detection performance among the different NMSE-NSDav combinations were found and the optimal trade-off from optimized reconstruction parameters corresponded to a maximum AUC value. The 4D MAP-RBI-EM with ADS correction, which had the best trade-off among the tested reconstruction methods, also had the highest AUC value, resulting in significantly better human observer detection performance when detecting regional myocardial wall motion abnormality. We concluded that the NMSE-NSDav trade-off was shown to agree with observer performance for the detection task of the regional motion abnormality, and the optimized 4D MAP-RBI-EM method with ADS corrections provides significant improvement compared to 3D FBP and 3D OS-EM with ADS corrections in detecting regional myocardial wall motion abnormali in 4D gated MP SPECT.

SU‐E‐J‐140: Simulation Study Using Thermoacoustics to Image Proton Dose and Range in Water and Skull Phantom

Purpose:In this study, thermoacoustic pressure signals generated from a proton beam were simulated in water and currently within a skull phantom to investigate the sensitivity of radioacoustic CT imaging in the brain.Methods:Thermoacoustically generated pressure signals from a pulse pencil proton beam (12, 15, 20, and 27cm range) were simulated in water. These simulated pressure signal are detected using a (71) transducer array placed along the surface of a cylinder (30cm × 40cm) and rotated over 2π (in 2 degree increments), where the normal vector to the surface of each transducer intersects the isocenter of the scanner. Currently, a software skull phantom is positioned at isocenter, where the scattering, absorption and speed of dispersion of the thermoacoustic signal through a three layer cortical‐trabecular‐cortical structure is being simulated. Based on data obtained from the literature, the effects of acoustic attenuation and speed‐of‐sound (dispersion) will be applied within the 3D FBP algorithm to obtain dosimetric images.Results:Based on hydrophone detector specifications, a 0.5MHz bandwidth and 50dB re 1μPa per Hz^1/2, a 1.6cGy sensitivity at the Bragg peak was demonstrated while maintaining a 1.0 mm (FWHM) range resolution along the central axis of the beam. Utilizing this same information, the integral dose within the Bragg peak and distal edge compared to MC had a 2% (statistical) and 5% voxel‐based RMS at this same dose sensitivity. We plan to present preliminary data determining the range sensitivity for a head phantom for this scanner design and the feasibility of imaging the proton dose in patients with a brain tumor undergoing therapy.Conclusion:RACT scanner provides 3D dosimetric images with 1.6cGy (Bragg peak) sensitivity with 1mm range sensitivity. Simulations will be performed to determine feasibility to treat brain cancer patients.

Soft-Tissue Imaging in Low-Dose, C-Arm Cone-Beam CT Using Statistical Image Reconstruction.

C-arm cone-beam CT (CBCT) is an emerging tool for intraoperative imaging, but current embodiments exhibit modest soft-tissue imaging capability and are largely constrained to high-contrast imaging tasks. A major advance in image quality is facilitated by statistical iterative reconstruction techniques. This work adapts a general penalized likelihood (PL) reconstruction approach with variable penalties and regularization to C-arm CBCT and investigates performance in imaging of large (>10 mm), low-contrast (<100 HU) tasks pertinent to soft-tissue surgical guidance. Experiments involved a mobile C-arm for CBCT with phantoms and cadavers presenting soft-tissue structures imaged using 3D filtered backprojection (FBP), quadratic, and non-quadratic PL reconstruction. Polyethylene phantoms with various tissue-equivalent inserts were used to quantity contrast-to-noise / resolution tradeoffs in low-contrast (~40 HU) structures, and the optimal reconstruction parameters were translated to imaging an anthropomorphic head phantom with low-contrasts targets and a cadaveric torso. Statistical reconstruction - especially non-quadratic PL variants - boosted soft-tissue image quality through reduction of noise and artifacts (e.g., a ~2-4 fold increase in contrast-to-noise ratio (CNR) at equivalent spatial resolution). For tasks relating to large, low-contrast tissues, even greater gains were possible using non-quadratic penalties and strong regularization that sacrificed spatial resolution in a manner still consistent with the imaging task. The advances in image quality offered by statistical reconstruction present promise and new challenges for interventional imaging, with high-speed computing facilitating realistic application. Careful investigation of performance relative to specific imaging tasks permits knowledgeable application of such techniques in a manner that overcomes conventional tradeoffs in noise, resolution, and dose and could extend application of CBCT-capable C-arms to soft-tissue interventions in neurosurgery as well as thoracic and abdominal interventions.

Image Quality and Quantitative Accuracy of Small Animal PET Images: Role of Physical Effects Correction and Reconstruction Algorithm

Positron emission tomography (PET) provides important metabolic information about tracer kinetics. Photon attenuation, scattered radiation, spatial resolution limitation and reconstruction algorithm parameters are variables that influence image quality and quantitative accuracy. The objective of this work was to examine the impact of the combined correction of photon attenuation and scatter on the quality and quantitative accuracy of small animal PET images reconstructed using the analytic filtered back projection algorithm (FBP) and the iterative maximum a posteriori probability (MAP) technique at different values of the smoothing parameter β.Methods. A homogenous water phantomandthree different configurations of NEMA image quality phantom were employed using a small animal PET scanner (InveonTM, Siemens medical solutions, Inc). A 20 min PET/CT scans were performed for all phantoms using a list mode acquisition. All Data were sorted into 3D sinograms and reconstructed using 2D FBP after Fourier rebinning and also reconstructed using OSEM3D/MAP (2 iteration OSEM3D and 18 iterations using MAP). Figures of merit used were error of tracer estimate, recovery coefficient, percentage standard deviation, spillover ratio and image noise. Results. Based on object geometry and size, FBP with attenuation and scatter correction provides accurate tracer concentration within the range of 5% and possible improvement can be achieved using MAP reconstruction. Without correction, errors of activity quantitation is significantly high, on average 34.5±7.2% Recovery coefficient was also found to improve with proper tuning of the β value using MAP reconstruction. Spillover ratio indicated some variations between FBP and MAP in air and water compartments, which might be attributed to the performance of the scatter correction. Noise measurements were superior with MAP reconstruction especially at high β values, but this occurs with edge artifacts and slight distortion of some geometric structures. Very small values of the beta have better resolution characteristics but high values have increased smoothing effect and blurring appearance. Conclusion: Attenuation correction is important in small animal PET and allows for reliable tracer estimate. Spatial resolution is improved by MAP technique and low β values provide high resolution capabilities and recovery coefficients but high values improve noise levels. For a given application, proper selection and employment of the smoothing parameter β should be followed. Scatter correction need to be improved especially in large objects or simultaneous acquisition of multiple numbers of animals.

The Evaluation of Corrective Reconstruction Method For Reduced Acquisition Time and Various Anatomies of Perfusion Defect Using Channelized Hotelling Observer for Myocardial Perfusion SPECT.

We evaluated the effect of conventional and corrective image reconstruction methods on reduced acquisition time for detecting a myocardial perfusion (MP) defect in MP SPECT using the Channelized Hotelling Observer (CHO). Using the 4D Extended Cardiac-Torso (XCAT) phantom, we simulated realistic transmural and endocardial MP defects in various location and size. Realistic Tc-99m Sestamibi MP projection data were generated using an analytical projector that included the effects of attenuation (A), collimator-detector response (D) and scatter (S) for various count levels simulating different acquisition times. They were reconstructed using the 3D FBP without correction and a 3D OS-EM method with ADS correction followed by a smoothing filter with various cut-off frequencies. The CHO followed by receiver operating characteristics (ROC) methodology was applied to the reconstructed images to evaluate the detectability of a MP defect in each method for different defect anatomies and count levels. Areas under the ROC curve (AUC) were computed to assess the changes in the MP defect detection. The results showed that the 3D OS-EM with ADS corrections showed significantly less changes in AUC value and gave overall higher AUC values than FBP at all cut-off frequencies of the post smoothing filter, count levels and MP defect sizes. The difference in AUC increased towards less smoothed images where the 3D OS-EM with correction was able to provide similar AUC values with 20 - 40% reduction in acquisition time compared to FBP. The AUC values for smaller MP defects were lower for both reconstruction methods with smaller differences. We concluded that the 3D OS-EM with ADS corrections provides higher performance in the MP defect detection task. It allows increased reduction of acquisition time without loss of MP defect detection in MP SPECT compared to the conventional FBP method especially towards less smoothed images.

DEVELOPMENT OF AN IMAGE RECONSTRUCTION METHOD FOR MICRON-CT USING PIXE

A prototype of micron-CT for biological research is being developed at Tohoku University. This micron-CT uses a point X-ray source with a spot size of 1μm and an X-ray CCD with 1000×1000 pixels of 8μm×8μm, achieving a spatial resolutions of the order of micro-meter. The event data obtained by the X-ray CCD is statistically poor and the 3 dimensional filtered back projection (3D FBP) algorithm, generally used in image reconstruction of X-ray CT, is not suitable because it is highly sensitive to statistical noise. Hence, we applied the expectation maximization (EM) algorithm for image reconstruction and developed an image reconstruction method using 3D EM algorithm. To confirm the validity of the reconstruction method, we irradiated two hairs inside a micro tube and reconstructed the CT image applying both EM and FBP algorithm on projection data. With 200×200×200 voxels of 4μm×4μm×4μm in the field of view, the computation time was less than 2 mins per iteration on a DELL Precision 650 Workstation 3.2GHz. The resulting EM image showed a better contrast than FBP image, and in the EM reconstructed CT image, we were able to reconstruct the micro tube of 270μm diameter and 45μm wall thickness and to visualize the two hairs inside.

Weighted FBP—a simple approximate 3D FBP algorithm for multislice spiral CT with good dose usage for arbitrary pitch

A new 3D reconstruction scheme, weighted filtered backprojection (WFBP) for multirow spiral CT based on an extension of the two-dimensional SMPR algorithm is described and results are presented. In contrast to other 3D algorithms available, the algorithm makes use of all available data for all pitch values. The algorithm is a FBP algorithm: linear convolution of the parallel data along the row direction followed by a 3D backprojection. Data usage for arbitrary pitch values is maintained through a weighting scheme which takes into account redundant data. If proper row weighting is applied, the image quality is superior to the image quality of the SMPR algorithm.

Three-dimensional diffraction tomography by two-dimensional sectioning

The hybrid filtered backpropagation (HFBP) algorithm of diffraction tomography is a reconstruction algorithm that provides quantitative reconstructions of weakly scattering objects from scattered field data collected in a suite of scattering experiments employing incident plane waves. Although the HFBP algorithm is capable of reconstructing three-dimensional (3D) objects it is most often used in applications involving 2D objects (cylinders) since the data acquisition and computational requirements for full 3D objects are quite severe. In this paper we examine the feasibility of using the 2D HFBP algorithm to reconstruct weakly scattering three-dimensional (3D) objects cross-section by cross-section, so as to obtain substantial savings in computation and data acquisition. The method is tested in a computer simulation where a sphere is reconstructed section by section using the 2D HFBP algorithm applied to an exactly computed scattered field obtained using an eigenfunction expansion. The (computed) total field (incident plus scattered) is first backpropagated to the centre of the image area from which the 2D FBP algorithm is used to reconstruct different sections of the sphere. We show that sections of non-attenuating spheres can be reconstructed to the same accuracy as for circular cylinders of the same radius, except for sections very close to the top or bottom of the spheres. For sections of highly attenuating spheres we do not obtain quite as good results as for the corresponding circular cylinders.

Characterization of a single LSO crystal layer High Resolution Research Tomograph

The purpose of this study was to determine the performance of a single lutetium oxy-orthosilicate (LSO) crystal layer High Resolution Research Tomograph (HRRT) positron emission tomography (PET) scanner. The HRRT is a high resolution PET scanner designed for human brain and small animal imaging. The scanner consists of eight panel detectors, which have one layer of 2.1 × 2.1 × 7.5 mm thick LSO crystals.Several phantom studies were performed to determine scanner characteristics, such as resolution, scatter fraction, count rate and noise equivalent count rates (NECR). NECR curves were measured according to both NEMA NU2-1994 and NU2-2001 for three different energy windows, i.e. lower level discriminators (lld) of 350, 400 and 450 keV and an upper level discriminator (uld) of 650 keV. Accuracy of scatter and single photon attenuation corrections was evaluated according to NU2-1994. Data were acquired using a ring difference of 67 and a span of 9. Reconstructions were performed using FORE + 2D FBP or OSEM.Transaxial resolution varied from 2.7 to 2.9 mm FWHM between 1 and 10 cm off centre locations, and axial resolution varied from 3.2 to 4.4 mm FWHM. Scatter fractions (NU2-1994) equalled 0.31, 0.42 and 0.54 for lld of 450, 400 and 350 keV, respectively. NECR data were highest for an lld of 400 keV and showed a maximum of 46 kcps at 38 kBq cm−3. Lower NECR values were observed according to NU2-2001, but were still optimal for an lld of 400 keV. After scatter and attenuation corrections, pixel values within water, air and teflon inserts of the NU2-1994 phantom were 14, 4 and 35% of the background activity, respectively.The single layer LSO HRRT scanner shows excellent spatial resolution, making it suitable for small animal studies. The low count rate performance, due to the small amount of LSO, prohibits studies of the human brain, but is sufficient for studies in small laboratory animals.

Comparison of FORE, OSEM and SAGE algorithms to 3DRP in 3D PET using phantom and human subject data

Using phantom and human subject data, the authors have compared a number of reconstruction algorithms to the current "gold standard" method, 3D Reprojection Method (3DRP), in an effort to validate them as practical 3D reconstruction alternatives for dynamic PET scanning. The algorithms evaluated were (a) Fourier Rebinning (FORE) followed by 2D Filtered Back Projection (2D FBP), (b) FORE followed by Ordered Subsets Expectation-Maximization (OSEM) and (c) FORE followed by Space Alternating Generalized Expectation-Maximization (SAGE). The main benefits of these methods are two-fold: significantly shorter reconstruction times and improved signal-to-noise ratio at matched resolution for the iterative schemes. The authors demonstrate that FORE+2D FBP can replace 3DRP with no significant impact on subsequent data analysis. In particular, distribution volume ratios (DVRs) obtained with FORE+2D FBP differed from those obtained with 3DRP by 0.3/spl plusmn/0.7% (n=33) and -0.7/spl plusmn/1.1% (n=27) for cortical and cerebellar input functions, respectively. Both OSEM and SAGE performed well compared to 3DRP, with improved noise characteristics and DVR differences of <3%.

Spline-based inverse Radon transform in two and three dimensions

While the exact inverse Radon transform is a continuous integral equation, the discrete nature of the data output by tomographic imaging systems generally demands that images be reconstructed using a discrete approximation to the transform. However, by fitting an analytic function to the projection data prior to reconstruction, one can avoid such approximations and preserve and exploit the continuous nature of the inverse transform. The authors present methods for the evaluation of the inverse Radon transform in two and three dimensions in which cubic spline functions are fit to the projection data, allowing the integrals that represent the filtration of the sinogram to be carried out in closed form and also eliminating the need for interpolation upon backprojection. Moreover, in the presence of noise, the algorithm can be used to reconstruct directly from the coefficients of smoothing splines, which are the minimizers of a popular curve-fitting measure. The authors find that the 2D and 3D direct-spline algorithms have superior resolution to their 2D and 3D FBP counterparts, albeit with higher noise levels, and that they have slightly lower ideal-observer signal-to-noise ratios for the detection of a 1-cm, spherical lesion with a 6:1 lesion-background concentration ratio.