James E. Bowsher

Bayesian reconstruction and use of anatomical a priori information for emission tomography

A Bayesian method is presented for simultaneously segmenting and reconstructing emission computed tomography (ECT) images and for incorporating high-resolution, anatomical information into those reconstructions. The anatomical information is often available from other imaging modalities such as computed tomography (CT) or magnetic resonance imaging (MRI). The Bayesian procedure models the ECT radiopharmaceutical distribution as consisting of regions, such that radiopharmaceutical activity is similar throughout each region. It estimates the number of regions, the mean activity of each region, and the region classification and mean activity of each voxel. Anatomical information is incorporated by assigning higher prior probabilities to ECT segmentations in which each ECT region stays within a single anatomical region. This approach is effective because anatomical tissue type often strongly influences radiopharmaceutical uptake. The Bayesian procedure is evaluated using physically acquired single-photon emission computed tomography (SPECT) projection data and MRI for the three-dimensional (3-D) Hoffman brain phantom. A clinically realistic count level is used. A cold lesion within the brain phantom is created during the SPECT scan but not during the MRI to demonstrate that the estimation procedure can detect ECT structure that is not present anatomically.

Fully Bayesian estimation of Gibbs hyperparameters for emission computed tomography data

In recent years, many investigators have proposed Gibbs prior models to regularize images reconstructed from emission computed tomography data. Unfortunately, hyperparameters used to specify Gibbs priors can greatly influence the degree of regularity imposed by such priors and, as a result, numerous procedures have been proposed to estimate hyperparameter values, from observed image data. Many of these, procedures attempt to maximize the joint posterior distribution on the image scene. To implement these methods, approximations to the joint posterior densities are required, because the dependence of the Gibbs partition function on the hyperparameter values is unknown. Here, the authors use recent results in Markov chain Monte Carlo (MCMC) sampling to estimate the relative values of Gibbs partition functions and using these values, sample from joint posterior distributions on image scenes. This allows for a fully Bayesian procedure which does not fix the hyperparameters at some estimated or specified value, but enables uncertainty about these values to be propagated through to the estimated intensities. The authors utilize realizations from the posterior distribution for determining credible regions for the intensity of the emission source. The authors consider two different Markov random field (MRF) models-the power model and a line-site model. As applications they estimate the posterior distribution of source intensities from computer simulated data as well as data collected from a physical single photon emission computed tomography (SPECT) phantom.

An EM algorithm for estimating SPECT emission and transmission parameters from emission data only

A maximum-likelihood (ML) expectation-maximization (EM) algorithm (called EM-IntraSPECT) is presented for simultaneously estimating single photon emission computed tomography (SPECT) emission and attenuation parameters from emission data alone. The algorithm uses the activity within the patient as transmission tomography sources, with which attenuation coefficients can he estimated. For this initial study, EM-IntraSPECT was tested on computer-simulated attenuation and emission maps representing a simplified human thorax as well as on SPECT data obtained from a physical phantom. Two evaluations were performed. First, to corroborate the idea of reconstructing attenuation parameters from emission data, attenuation parameters (/spl mu/) were estimated with the emission intensities (/spl lambda/) fixed at their true values. Accurate reconstructions of attenuation parameters were obtained. Second, emission parameters /spl lambda/ and attenuation parameters Cl were simultaneously estimated from the emission data alone. In this case there was crosstalk between estimates of /spl lambda/ and /spl mu/ and final estimates of /spl lambda/ and /spl mu/ depended on initial values. Estimates degraded significantly as the support extended out farther from the body, and an explanation for this is proposed. In the EM-IntraSPECT reconstructed attenuation images, the lungs, spine, and soft tissue were readily distinguished and had approximately correct shapes and sizes. As compared with standard EM reconstruction assuming a fix uniform attenuation map, EM-IntraSPECT-provided more uniform estimates of cardiac activity in the physical phantom study and in the simulation study with tight support, but less uniform estimates with a broad support. The new EM algorithm derived here has additional applications, including reconstructing emission and transmission projection data under a unified statistical model.

A 3D gantry single photon emission tomograph with hemispherical coverage for dedicated breast imaging

Abstract A novel tomographic gantry was designed, built and initially evaluated for single photon emission imaging of metabolically active lesions in the pendant breast and near chest wall. Initial emission imaging measurements with breast lesions of various uptake ratios are presented. Methods: A prototype tomograph was constructed utilizing a compact gamma camera having a field-of-view of Results: As iteration number increased for the tomographically measured data at all polar angles, contrasts increased while signal-to-noise ratios (SNRs) decreased in the expected way with OSEM reconstruction. The rollover between contrast improvement and SNR degradation of the lesion occurred at two to three iterations. The reconstructed tomographic data yielded SNRs with or without scatter correction that were >9 times better than the planar scans. There was up to a factor of ∼2.5 increase in total primary and scatter contamination in the photopeak window with increasing tilt angle from 15° to 45°, consistent with more direct line-of-sight of myocardial and liver activity with increased camera polar angle. Conclusion: This new, ultra-compact, dedicated tomographic imaging system has the potential of providing valuable, fully 3D functional information about small, otherwise indeterminate breast lesions as an adjunct to diagnostic mammography.

Evaluation of fully 3-D emission mammotomography with a compact cadmium zinc telluride detector

A compact, dedicated cadmium zinc telluride (CZT) gamma camera coupled with a fully three-dimensional (3-D) acquisition system may serve as a secondary diagnostic tool for volumetric molecular imaging of breast cancers, particularly in cases when mammographic findings are inconclusive. The developed emission mammotomography system comprises a medium field-of-view, quantized CZT detector and 3-D positioning gantry. The intrinsic energy resolution, sensitivity and spatial resolution of the detector are evaluated with Tc-99m (140 keV) filled flood sources, capillary line sources, and a 3-D frequency-resolution phantom. To mimic realistic human pendant, uncompressed breast imaging, two different phantom shapes of an average sized breast, and three different lesion diameters are imaged to evaluate the system for 3-D mammotomography. Acquisition orbits not possible with conventional emission, or transmission, systems are designed to optimize the viewable breast volume while improving sampling of the breast and anterior chest wall. Complications in camera positioning about the patient necessitate a compromise in these two orbit design criteria. Image quality is evaluated with signal-to-noise ratios and contrasts of the lesions, both with and without additional torso phantom background. Reconstructed results indicate that 3-D mammotomography, incorporating a compact CZT detector, is a promising, dedicated breast imaging technique for visualization of tumors <1 cm in diameter. Additionally, there are no outstanding trajectories that consistently yield optimized quantitative lesion imaging parameters. Qualitatively, imaging breasts with realistic torso backgrounds (out-of-field activity) substantially alters image characteristics and breast morphology unless orbits which improve sampling are utilized. In practice, the sampling requirement may be less strict than initially anticipated.

Utilizing MRI information to estimate F18-FDG distributions in rat flank tumors

This paper investigates the potential of magnetic resonance imaging (MRI) to improve the estimation of within-tumor variations in F18-FDG concentration. An image model is described for incorporating MRI images into positron emission tomography (PET) and single photon emission computed tomography (SPECT) radiotracer image reconstruction. The model promotes greater smoothing, of estimated radiotracer concentration, among nearby voxels that have more nearly similar MRI signals. R3230 mammary adenocarcinomas are grown on rat flanks. Autoradiography, histology, and T2-weighted MRI are used to demonstrate that the above image model accurately reflects true F18-FDG distributions in R3230 tumors. In vivo F18-FDG distributions are then reconstructed from PET projection data, with and without incorporating MRI. The F18-FDG images reconstructed with MRI show greater detail, and this additional detail is consistent with the results of the autoradiography and histology studies.

Dosimetric characteristics of Novalis Tx system with high definition multileaf collimator

A new Novalis Tx system equipped with a high definition multileaf collimator (HDMLC) recently became available to perform both image-guided radiosurgery and conventional radiotherapy. It is capable of delivering a highly conformal radiation dose with three energy modes: 6 MV photon energy, 15 MV photon energy, and 6 MV photon energy in a stereotactic radiosurgery mode with 1000 MU/min dose rate. Dosimetric characteristics of the new Novalis Tx treatment unit with the HDMLC are systematically measured for commissioning. A high resolution diode detector and miniion-chamber detector are used to measure dosimetric data for a range of field sizes from 4 x 4 mm to 400 x 400 mm. The commissioned Novalis Tx system has passed the RPC stereotactic radiosurgery head phantom irradiation test. The Novalis Tx system not only expands its capabilities with three energy modes, but also achieves better beam conformity and sharer beam penumbra with HDMLC. Since there is little beam data information available for the new Novalis Tx system, we present in this work the dosimetric data of the new modality for reference and comparison.

Commissioning and dosimetric characteristics of TrueBeam system: Composite data of three TrueBeam machines

PURPOSE A TrueBeam linear accelerator (TB-LINAC) is designed to deliver traditionally flattened and flattening-filter-free (FFF) beams. Although it has been widely adopted in many clinics for patient treatment, limited information is available related to commissioning of this type of machine. In this work, commissioning data of three units were measured, and multiunit comparison was presented to provide valuable insights and reliable evaluations on the characteristics of the new treatment system. METHODS The TB-LINAC is equipped with newly designed waveguide, carousel assembly, monitoring control, and integrated imaging systems. Each machine in this study has 4, 6, 8, 10, 15 MV flattened photon beams, and 6 MV and 10 MV FFF photon beams as well as 6, 9, 12, 16, 20, and 22 MeV electron beams. Dosimetric characteristics of the three new TB-LINAC treatment units are systematically measured for commissioning. High-resolution diode detectors and ion chambers were used to measure dosimetric data for a range of field sizes from 10 × 10 to 400 × 400 mm(2). The composite dosimetric data of the three units are presented in this work. The commissioning of intensity modulated radiotherapy (IMRT), volumetric modulated arc therapy (VMAT), image-guided radiation therapy, and gating systems are also illustrated. Critical considerations of P(ion) of FFF photon beams and small field dosimetric measurements were investigated. RESULTS The authors found all PDDs and profiles matched well among the three machines. Beam data were quantitatively compared and combined through average to yield composite beam data. The discrepancies among the machines were quantified using standard deviation (SD). The mean SD of the PDDs among the three units is 0.12%, and the mean SD of the profiles is 0.40% for 10 MV FFF open fields. The variations of P(ion) of the chamber CC13 is 1.2 ± 0.1% under 6 MV FFF and 2.0 ± 0.5% under 10 MV FFF from dmax to the 18 cm-off-axis point at 35 cm depth under 40 × 40 cm(2). The mean penumbra of crossplane flattened photon beams at collimator angle of 0° is measured from 5.88 ± 0.09 to 5.99 ± 0.13 mm from 4 to 15 MV at 10 cm depth of 100 × 100 mm(2). The mean penumbra of crossplane beams at collimator angle of 0° is measured as 3.70 ± 0.21 and 4.83 ± 0.04 mm for 6 MV FFF and 10 MV FFF, respectively, at 10 cm depth with a field size of 5 × 5 cm(2). The end-to-end test procedures of both IMRT and VMAT were performed for various energy modes. The mean ion chamber measurements of three units showed less than 2% between measurement and calculation; the mean MultiCube ICA measurements demonstrated over 90% pixels passing gamma analysis (3%, 3 mm, 5% threshold). The imaging dosimetric data of KV planar imaging and CBCT demonstrated improved consistency with vendor specifications and dose reduction for certain imaging protocols. The gated output verification showed a discrepancy of 0.05% or less between gating radiation delivery and nongating radiation delivery. CONCLUSIONS The commissioning data indicated good consistency among the three TB-LINAC units. The commissioning data provided us valuable insights and reliable evaluations on the characteristics of the new treatment system. The systematically measured data might be useful for future reference.

Design and development of a fully 3D dedicated x-ray computed mammotomography system

Our effort to implement a volumetric x-ray computed mammotomography (CmT) system dedicated to imaging breast disease comprises: demonstrated development of a quasi-monochromatic x-ray beam providing minimal dose and other optimal imaging figures of merit; new development of a compact, variable field-of-view, fully-3D acquisition gantry with a digital flat-panel detector facilitating more nearly complete sampling of frequency space and the physical breast volume; incorporation of iterative ordered-subsets transmission (OSTR) image reconstruction allowing modeling of the system matrix. Here, we describe the prototype 3D gantry and demonstrate initial system performance. Data collected on the prototype gantry demonstrate the feasibility of using OSTR with realistic reconstruction times. The gantry consists of a rotating W-anode x-ray tube using ultra-thick K-edge filtration, and an ~20x25cm2 digital flat-panel detector located at <60cm SID. This source/detector combination can be shifted laterally changing the location of the central ray relative to the system center-of-rotation, hence changing the effective imaging field-of-view, and is mounted on a goniometric cradle allowing <50° polar tilt, then on a 360° azimuthal rotation stage. Combined, these stages provide for positioning flexibility in a banded region about a sphere, facilitating simple circle-plus-arc-like trajectories, as well as considerably more complex 3D trajectories. Complex orbits are necessary to avoid physical hindrances from the patient while acquiring the largest imaging volume of the breast. The system capabilities are demonstrated with fully-3D reconstructed images of geometric sampling and resolution phantoms, a fabricated breast phantom containing internal features of interest, and a cadaveric breast specimen. This compact prototype provides flexibility in dedicated, fully-3D CmT imaging of healthy and diseased breasts.

Implementation and initial characterization of acquisition orbits with a dedicated emission mammotomograph

While contemporary clinical single photon emission computed tomography (SPECT) scanners are limited to a horizontal axis of rotation about the patient, the compact application specific emission tomography (ASET) system overcomes physical proximity limitations of clinical scanners and allows for fully three-dimensional movement in an inverted hemisphere about the pendant breast through combined variable radius of rotation (ROR) and polar and azimuthal angular positioning. With these three degrees of freedom, the ASET can provide trajectories that satisfy Orlovs sampling criterion while maintaining a small ROR, necessary to minimize resolution degradation. One class of orbits investigated here consists of combinations of circular orbits and up to 60/spl deg/ arcs. Orbits are evaluated for image quality, including sampling and resolution characterization, utilizing both cold disk and cold rod emission phantoms, and quantitated contrasts and signal-to-noise ratios (SNRs) of small lesions located in a breast phantom with and without additional torso phantom backgrounds. The reconstructed cold disk and cold rod results indicate that all tested orbits in this class completely sample the volume and provide near equal resolution recovery. Furthermore, results from all orbits yielded higher contrasts and/or SNRs for the breast lesions compared with simple vertical-axis-of-rotation acquisitions, and nearly an order of magnitude better than those from uncompressed planar imaging.