Stephen J. Glick

GATE: a simulation toolkit for PET and SPECT.

Monte Carlo simulation is an essential tool in emission tomography that can assist in the design of new medical imaging devices, the optimization of acquisition protocols and the development or assessment of image reconstruction algorithms and correction techniques. GATE, the Geant4 Application for Tomographic Emission, encapsulates the Geant4 libraries to achieve a modular, versatile, scripted simulation toolkit adapted to the field of nuclear medicine. In particular, GATE allows the description of time-dependent phenomena such as source or detector movement, and source decay kinetics. This feature makes it possible to simulate time curves under realistic acquisition conditions and to test dynamic reconstruction algorithms. This paper gives a detailed description of the design and development of GATE by the OpenGATE collaboration, whose continuing objective is to improve, document and validate GATE by simulating commercially available imaging systems for PET and SPECT. Large effort is also invested in the ability and the flexibility to model novel detection systems or systems still under design. A public release of GATE licensed under the GNU Lesser General Public License can be downloaded at http:/www-lphe.epfl.ch/GATE/. Two benchmarks developed for PET and SPECT to test the installation of GATE and to serve as a tutorial for the users are presented. Extensive validation of the GATE simulation platform has been started, comparing simulations and measurements on commercially available acquisition systems. References to those results are listed. The future prospects towards the gridification of GATE and its extension to other domains such as dosimetry are also discussed.

Noniterative compensation for the distance-dependent detector response and photon attenuation in SPECT imaging

A filtering approach is described, which accurately compensates for the 2D distance-dependent detector response, as well as for photon attenuation in a uniform attenuating medium. The filtering method is based on the frequency distance principle (FDP) which states that points in the object at a specific source-to-detector distance provide the most significant contribution to specified frequency regions in the discrete Fourier transform (DFT) of the sinogram. By modeling the detector point spread function as a 2D Gaussian function whose width is dependent on the source-to-detector distance, a spatially variant inverse filter can be computed and applied to the 3D DFT of the set of all sinogram slices. To minimize noise amplification the inverse filter is rolled off at high frequencies by using a previously published Wiener filter strategy. Attenuation compensation is performed with Bellinis method. It was observed that the tomographic point response, after distance-dependent filtering with the FDP, was approximately isotropic and varied substantially less with position than that obtained with other correction methods. Furthermore, it was shown that processing with this filtering technique provides reconstructions with minimal degradation in image fidelity.

A computer simulation study comparing lesion detection accuracy with digital mammography, breast tomosynthesis, and cone‐beam CT breast imaging

Although conventional mammography is currently the best modality to detect early breast cancer, it is limited in that the recorded image represents the superposition of a three-dimensional (3D) object onto a 2D plane. Recently, two promising approaches for 3D volumetric breast imaging have been proposed, breast tomosynthesis (BT) and CT breast imaging (CTBI). To investigate possible improvements in lesion detection accuracy with either breast tomosynthesis or CT breast imaging as compared to digital mammography (DM), a computer simulation study was conducted using simulated lesions embedded into a structured 3D breast model. The computer simulation realistically modeled x-ray transport through a breast model, as well as the signal and noise propagation through a CsI based flat-panel imager. Polyenergetic x-ray spectra of Mo/Mo 28 kVp for digital mammography, Mo/Rh 28 kVp for BT, and W/Ce 50 kVp for CTBI were modeled. For the CTBI simulation, the intensity of the x-ray spectra for each projection view was determined so as to provide a total average glandular dose of 4 mGy, which is approximately equivalent to that given in conventional two-view screening mammography. The same total dose was modeled for both the DM and BT simulations. Irregular lesions were simulated by using a stochastic growth algorithm providing lesions with an effective diameter of 5 mm. Breast tissue was simulated by generating an ensemble of backgrounds with a power law spectrum, with the composition of 50% fibroglandular and 50% adipose tissue. To evaluate lesion detection accuracy, a receiver operating characteristic (ROC) study was performed with five observers reading an ensemble of images for each case. The average area under the ROC curves (Az) was 0.76 for DM, 0.93 for BT, and 0.94 for CTBI. Results indicated that for the same dose, a 5 mm lesion embedded in a structured breast phantom was detected by the two volumetric breast imaging systems, BT and CTBI, with statistically significant higher confidence than with planar digital mammography, while the difference in lesion detection between BT and CTBI was not statistically significant.

Comparison of tomosynthesis methods used with digital mammography

RATIONALE AND OBJECTIVES The authors performed this study to investigate the potential applicability of tomosynthesis to digital mammography. Four methods of tomosynthesis-tuned aperture computed tomography (TACT)-backprojection, TACT-iterative restoration, iterative reconstruction with expectation maximization, and Bayesian smoothing-were compared to planar mammography and analyzed in terms of their contrast-detail characteristics. Specific comparisons between the tomosynthesis methods were not attempted in this study. MATERIALS AND METHODS A full-field, amorphous, silicon-based, flat-panel digital mammographic system was used to obtain planar and tomosynthesis projection images. A composite tomosynthesis phantom with a centrally located contrast-detail insert was used as the object of interest. The total exposure for multiple views with tomosynthesis was always equal to or less than that for the planar technique. Algorithms were used to reconstruct the object from the acquired projections. RESULTS Threshold contrast characteristics with all tomosynthesis reconstruction methods were significantly better than those with planar mammography, even when planar mammography was performed at more than twice the exposure level. Reduction of out-of-plane structural components was observed in all the tomosynthesis methods analyzed. CONCLUSION The contrast-detail trends of all the tomosynthesis methods analyzed in this study were better than those of planar mammography. Further optimization of the algorithms could lead to better image reconstruction, which would improve visualization of valuable diagnostic information.

Attenuation compensation for cardiac single-photon emission computed tomographic imaging: Part 2. Attenuation compensation algorithms☆☆☆

Attenuation is believed to be one of the major causes of false-positive cardiac single-photon emission computed tomographic perfusion images. This article provides an introduction to the approaches used to correct for nonuniform attenuation once a patient-specific attenuation map is available. Comparison is made of specific attenuation-correction algorithms from each of three major categories of compensation methods that are or will be available commercially. Examples of the use of the algorithms on simulated projections of a mathematic phantom modeling the anatomy of the upper torso are used to illustrate the ability of the methods to compensate for attenuation. The advantages and disadvantages of each approach are summarized, as well as areas that need further investigation.

Noise characterization of block-iterative reconstruction algorithms. I. Theory

Researchers have shown increasing interest in block-iterative image reconstruction algorithms due to the computational and modeling advantages they provide. Although their convergence properties have been well documented, little is known about how they behave in the presence of noise. In this work, the authors fully characterize the ensemble statistical properties of the rescaled block-iterative expectation-maximization (RBI-EM) reconstruction algorithm and the rescaled block-iterative simultaneous multiplicative algebraic reconstruction technique (RBI-SMART). Also included in the analysis are the special cases of RBI-EM, maximum-likelihood EM (ML-EM) and ordered-subset EM (OS-EM), and the special case of RBI-SMART, SMART. A theoretical formulation strategy similar to that previously outlined for ML-EM is followed for the RBI methods. The theoretical formulations in this paper rely on one approximation, namely, that the noise in the reconstructed image is small compared to the mean image. In a second paper, the approximation will be justified through Monte Carlo simulations covering a range of noise levels, iteration points, and subset orderings. The ensemble statistical parameters could then be used to evaluate objective measures of image quality.

Microcalcification detection using cone-beam CT mammography with a flat-panel imager

The purpose of this study was to investigate microcalcification detectability using CT mammography with a flat-panel imager. To achieve this, a computer simulation was developed to model an amorphous-silicon, CsI based flat-panel imager system using a linear cascaded model. The breast was modelled as a hemi-ellipsoid shape with composition of 50% adipose and 50% glandular tissue. Microcalcifications were modelled as small spheres having a composition of calcium carbonate. The results show that with a mean glandular dose equivalent to that typically used in two-view screening mammography, CT mammography with a flat-panel detector is capable of providing images where most microcalcifications are detectable. A receiver operating characteristic (ROC) study was conducted by five physicist observers viewing simulated CT mammography reconstructions. The results suggest that the microcalcification with its diameter equal to or greater than 0.175 mm can be detected with an average area under the ROC curve (AUC) greater than 0.95 using 0.1 or 0.2 mm pixelized detectors. The results also indicate that the optimal pixel size of the detector is around 0.2 mm for microcalcification detection, based on the trade-off between detectability of microcalcifications and the time required for data acquisition and reconstruction.

Reducing the influence of the partial volume effect on SPECT activity quantitation with 3D modelling of spatial resolution in iterative reconstruction

Quantitative parameters such as the maximum and total counts in a volume are influenced by the partial volume effect. The magnitude of this effect varies with the non-stationary and anisotropic spatial resolution in SPECT slices. The objective of this investigation was to determine whether iterative reconstruction which includes modelling of the three-dimensional (3D) spatial resolution of SPECT imaging can reduce the impact of the partial volume effect on the quantitation of activity compared with filtered backprojection (FBP) techniques which include low-pass, and linear restoration filtering using the frequency distance relationship (FDR). The iterative reconstruction algorithms investigated were maximum-likelihood expectation-maximization (MLEM), MLEM with ordered subset acceleration (ML-OS), and MLEM with acceleration by the rescaled-block-iterative technique (ML-RBI). The SIMIND Monte Carlo code was used to simulate small hot spherical objects in an elliptical cylinder with and without uniform background activity as imaged by a low-energy ultra-high-resolution (LEUHR) collimator. Centre count ratios (CCRs) and total count ratios (TCRs) were determined as the observed counts over true counts. CCRs were unstable while TCRs had a bias of approximately 10% for all iterative techniques. The variance in the TCRs for ML-OS and ML-RBI was clearly elevated over that of MLEM, with ML-RBI having the smaller elevation. TCRs obtained with FDR-Wiener filtering had a larger bias (approximately 30%) than any of the iterative reconstruction methods but near stationarity is also reached. Butterworth filtered results varied by 9.7% from the centre to the edge. The addition of background has an influence on the convergence rate and noise properties of iterative techniques.

Restoration of combined conjugate images in SPECT: comparison of a new Wiener filter and the image-dependent Metz filter

Two image-dependent restoration filters were applied to projection image sets obtained with single-photon-emission computed tomography (SPECT). Wiener and Metz restorations of combined conjugate views are compared to each other, Wiener restoration of individual projection images, and one-dimensional Butterworth smoothing. The combined view restoration filters adapt to the average thickness of the object by estimating a modulation transfer function (MTF) for that thickness. Simulated Tc-99m liver-spleen studies with randomly placed cold spot tumors, a projector which accounts for the spatially variant blurring in SPECT, and a Poisson noise generator are used to compute simulated projection image sets. These sets are filtered and reconstructed using the method of intrinsic attenuation correction of Tanaka et al. (1984). Cold spot contrast, automated receiver operator characteristic (ROC) analysis of cold spot detectability, and the normalized mean squared error (NMSE) are used to compare the four processing methods. Little difference is noted between the three restoration methods. However, the restoration filters all yield noticeably better ROC and NMSE results than the one-dimensional smoothing. Inspection of the three-dimensional MTF derived from reconstructions of point sources indicates that the restoration filters substantially reduce the low-frequency degradation which is primarily due to scatter. >

Characterization of scatter in cone‐beam CT breast imaging: Comparison of experimental measurements and Monte Carlo simulation

It is commonly understood that scattered radiation in x-ray computed tomography (CT) degrades the reconstructed image. As a precursor to developing scatter compensation methods, it is important to characterize this scatter using both empirical measurements and Monte Carlo simulations. Previous studies characterizing scatter using both experimental measurements and Monte Carlo simulations have been reported in diagnostic radiology and conventional mammography. The emerging technology of cone-beam CT breast imaging (CTBI) differs significantly from conventional mammography in the breast shape and imaging geometry, aspects that are important factors impacting the measured scatter. This study used a bench-top cone-beam CTBI system with an indirect flat-panel detector. A cylindrical phantom with equivalent composition of 50% fibroglandular and 50% adipose tissues was used, and scatter distributions were measured by beam stop and aperture methods. The GEANT4-based simulation package GATE was used to model x-ray photon interactions in the phantom and detector. Scatter to primary ratio (SPR) measurements using both the beam stop and aperture methods were consistent within 5% after subtraction of nonbreast scatter contributions and agree with the low energy electromagnetic model simulation in GATE. The validated simulation model was used to characterize the SPR in different CTBI conditions. In addition, a realistic, digital breast phantom was simulated to determine the characteristics of various scatter components that cannot be separated in measurements. The simulation showed that the scatter distribution from multiple Compton and Rayleigh scatterings, as well as from the single Compton scattering, has predominantly low-frequency characteristics. The single Rayleigh scatter was observed to be the primary contribution to the spatially variant scatter component.