Slit-slat Collimator Research Articles

A novel reconstruction method combining multi-detector SPECT with an elliptical orbit and computer tomography for cardiac imaging

The myocardial single photon emission computed tomography (SPECT) is a good study due to its clinical significance in the diagnosis of myocardial disease and the requirement for improving image quality. However, SPECT imaging faces challenges related to low spatial resolution and significant statistical noise, which concerns patient radiation safety. In this paper, a novel reconstruction system combining multi-detector elliptical SPECT (ME-SPECT) and computer tomography (CT) is proposed to enhance spatial resolution and sensitivity. The hybrid imaging system utilizes a slit-slat collimator and elliptical orbit to improve sensitivity and signal-to-noise ratio (SNR), obtains accurate attenuation mapping matrices, and requires prior information from integrated CT. Collimator parameters are corrected based on CT reconstruction results. The SPECT imaging system employs an iterative reconstruction algorithm that utilizes prior knowledge. An iterative reconstruction algorithm based on prior knowledge is applied to the SPECT imaging system, and a method for prioritizing the reconstruction of regions of interest (ROI) is introduced to deal with severely truncated data from ME-SPECT. Simulation results show that the proposed method can significantly improve the system's spatial resolution, SNR, and image fidelity. The proposed method can effectively suppress distortion and artifacts with the higher spatial resolution ordered subsets expectation maximization (OSEM); slit-slat collimation.

A Statistical DOI Estimation Algorithm for a SiPM-Based Clinical SPECT Insert

A prototype clinical brain single-photon emission computed tomography (SPECT) insert has been designed for use in simultaneous SPECT/MRI. The system utilizes novel slit-slat collimators which, like pinhole collimators, suffers from parallax errors due to the large incident angle of photons. A statistical algorithm has been developed to determine the depth-of-interaction (DOI) with a view to improving image performance. The importance of DOI correction was demonstrated using Monte Carlo simulation. This simulation also indicated that four DOI layers ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3 \times 1.5\, \textrm {mm} + 3.5\, \textrm {mm}$ </tex-math></inline-formula> ) may be sufficient. The improvement in event localization was demonstrated on a single detector before implementing the algorithm on the full clinical prototype where some limitations in event localization in layers close to the readout plane were observed. Nevertheless, DOI enabled the rejection of poorly localized events with improved resolution in reconstructed line sources.

Review on Recent Developments in Collimators of Single Photon Emission Computed Tomography Imaging

In Single Photon Emission Computed Tomography (SPECT), collimator selection, optimization, and also geometric calibration have a major impact on the acquired image quality and also on an accurate detectability and diagnosis. The collimator optimization phenomena consider some parameters such as field of view, resolution, sensitivity, resolution at depth, septal thickness and penetration for a specific application task. While the parallel hole collimator is usually used in SPECT and planar imaging but due to the limited solid angle covered by the collimator, the system sensitivity and resolution were highly reduced. Meanwhile, other types of collimators such as pin-hole, multi-pin-hole, slant and slit-slat collimators were introduced with a trade-off between sensitivity and resolution. This article reviews improvements on collimators also by considering the geometry and geometric calibration methods for improving the image quality in single photon emission computed tomography.

A novel cardiac SPECT system and imaging method

ABSTRACT A novel SPECT model with an elliptical orbit and multiple detectors is proposed for detecting the metabolism of the human heart. Under the structure of the proposed model, the distance between the detector and the patient becomes smaller, which can effectively improve the SNR. The slit-slat collimator is used to collimate the emitted photons. Modified iterative reconstruction algorithm is proposed to reconstruct a small centre field with severe truncation data under the constraints of body contour and total variation (TV). Simulations for the proposed SPECT are implemented to access the image reconstruction quality. The simulation results show that proposed imaging system with appropriate reconstruction methods can provide significant improvements in the spatial resolution and the image quality. The geometric efficiency of the proposed SPECT is about twice that of the conventional SPECT system, but the scanning time is only about one tenth.

Hybrid Cardiac Imaging Integrating Elliptical Orbit SPECT With CT

The cardiac single photon emission computed tomography (SPECT) is an important research topic because of the large clinical demand and the need for improving image quality. It usually suffers from low count sensitivity and poor spatial resolution due to patient radiation safety concerns. In this study, hybrid cardiac imaging integrating elliptical orbit SPECT (E-SPECT) with computed tomography (CT) is proposed to better obtain the anatomical and physiological information of the cardiac and vascular systems. The geometric structure of this proposed hybrid imaging system is similar to that of an ellipse, containing multiple detectors with slit-slat collimation. The hybrid imaging system utilizes the elliptical geometry and multi-detectors to improve the sensitivity and signal to noise ratio (SNR), and incorporates a CT scanner into the system to obtain an accurate attenuation mapping matrix of the patient for image reconstruction. The iterative reconstruction algorithm with some constraints is used to reconstruct images. The constraint-based body contour is introduced to deal with severely truncated data from E-SPECT. The feasibility and advantage of the system is validated by simulations. Simulation results show that it is optimal for system design and the quality of reconstructed images are greatly improved. With an optimal sampling coverage angle between 210° and 240°, the hybrid cardiac imaging system proposed in the paper can achieves significant improvements in image spatial resolution, SNR and image fidelity.

Development of a Practical Calibration Procedure for a Clinical SPECT/MRI System Using a Single INSERT Prototype Detector and Multimini Slit-Slat Collimator

In the context of the INSERT project, we have been developing a clinical single photon emission computed tomography (SPECT) insert for a magnetic resonance imaging (MRI) system, in order to perform simultaneous SPECT/MRI of the human brain. This system will consist of 20 CsI:Tl scintillation detectors, 5-cm wide and 10-cm long, with a 72-channel silicon photomultiplier (SiPM) readout per detector, and a multimini slit-slat (MSS) collimator set up in a stationary partial ring. Additionally the system has a custom-built transmit/receive MR coil to ensure compatibility with the SPECT system. Due to the novel design of the system/collimator, existing geometric calibration methods are not suitable. Therefore we propose a novel and practical calibration procedure that consists of a set of specific independent measurements to determine the geometric parameters of the collimator. This procedure was developed utilizing a prototype system that consists of a reduced-size single detector with a 36-channel SiPM-based readout and a single MSS collimator module. Validation was performed by reconstructing different imaging phantoms, using a rotating stage to simulate a tomographic acquisition. Regarding uniformity, the coefficient of variation (COV) for the cylinder phantom reconstructed with correct calibration parameters is 6.7%, whereas the COV using incorrect parameters is 9.4%. The quality of the phantom reconstructions provide evidence of the applicability of the proposed method to the calibration of the prototype system. This procedure can be easily adapted for the final INSERT system.

Comparison of fan beam, slit-slat and multi-pinhole collimators for molecular breast tomosynthesis

Recently, we proposed and optimized dedicated multi-pinhole molecular breast tomosynthesis (MBT) that images a lightly compressed breast. As MBT may also be performed with other types of collimators, the aim of this paper is to optimize MBT with fan beam and slit-slat collimators and to compare its performance to that of multi-pinhole MBT to arrive at a truly optimized design. Using analytical expressions, we first optimized fan beam and slit-slat collimator parameters to reach maximum sensitivity at a series of given system resolutions. Additionally, we performed full system simulations of a breast phantom containing several tumours for the optimized designs. We found that at equal system resolution the maximum achievable sensitivity increases from pinhole to slit-slat to fan beam collimation with fan beam and slit-slat MBT having on average a 48% and 20% higher sensitivity than multi-pinhole MBT. Furthermore, by inspecting simulated images and applying a tumour-to-background contrast-to-noise (TB-CNR) analysis, we found that slit-slat collimators underperform with respect to the other collimator types. The fan beam collimators obtained a similar TB-CNR as the pinhole collimators, but the optimum was reached at different system resolutions. For fan beam collimators, a 6–8 mm system resolution was optimal in terms of TB-CNR, while with pinhole collimation highest TB-CNR was reached in the 7–10 mm range.

Collimator Design for a Brain SPECT/MRI Insert

This project’s goal is to design a SPECT insert for a clinical MRI system for simultaneous brain SPECT/MR imaging, with a high-sensitivity collimator and high-resolution detectors. We have compared eight collimator designs, four multi-pinhole and four multi-slit slit-slat configurations. The collimation was designed for a system with rings of 25 $5\times 5~\hbox{cm}$ detectors. We introduce the concept of 1/2-pinhole and 1/2-slit, which are transaxially shared between two adjacent detectors. Analytical geometric efficiency was calculated for an activity distribution corresponding to a human brain and a range of intrinsic detector resolutions ${\rm R}_i$ and target resolutions ${\rm R}_t$ at the centre of the FOV. Noise-free data were simulated with and without depth-of-interaction (DOI) information, 0.8 mm ${\rm R}_i$ and 10 mm ${\rm R}_t$ FWHM, and reconstructed for uniform, Defrise, Derenzo, and Zubal brain phantoms. Comparing the multi-pinhole and multi-slit slit-slat collimators, the former gives better reconstructed uniformity and transaxial resolution, while the latter gives better axial resolution. Although the $2 \times 2$ -pinhole and 2-slit designs give the highest sensitivities, they result in a sub-optimal utilisation of the detector FOV. The best options are therefore the $ 5+ 2 1/2$ -pinhole and the $1 + 2 1/2$ -slit systems, with sensitivities of $1.8 \times {10^{ - 4}}$ and $3.2 \times {10^{ - 4}}$ , respectively. Noiseless brain phantom reconstructions with the multi-pinhole collimator are slightly superior as compared to slit-slat, in terms of symmetry and accuracy of the activity distribution, but the same is not true when noise is included. DOI information reduces artefacts and improves uniformity in geometric phantoms. Further evaluation is needed with prototype collimators.

Collimator design for a clinical brain SPECT/MRI insert.

This project's goal is to design a SPECT insert for a clinical MRI system for simultaneous brain SPECT/MR imaging. We assume the stationary SPECT insert will consist of two rings of ∼5x5-cm SiPM-based detectors insensitive to magnetic fields, with 0.8-mm intrinsic resolution. The maximum diameter is 44.5 cm, the minimum diameter is 33 cm to accommodate the patient and MRI receive/transmit coil, and the FOV has a 20 cm diameter. We have compared eight collimator designs: single-, 2x2-, 3x3- and 5+2½- pinhole, and single-, 2-, 3- and 1+2½-slit slit-slat, where ½-pinholes/slits are shared between two detectors. Analytical geometric efficiency was calculated for an activity distribution corresponding to a human brain and a target resolution of 10 mm FWHM at the centre of the FOV. Noise-free data were simulated with and without depth-of-interaction (DOI) information, and reconstructed for uniform, Defrise, Derenzo, and Zubal brain phantoms. For DOI it is assumed that the crystal's first and second half can be differentiated. Comparing the multi-pinhole and multi-slit slit-slat collimators, the former gives better reconstructed uniformity and trans-axial resolution, while the latter gives better axial resolution. Although the 2x2-pinhole and 2-slit designs give the highest sensitivities, they result in a sub-optimal utilization of the detector FOV. The best options are therefore the 5+2½-pinhole and the 1+2½-slit systems, with sensitivities of 4.9*10–4 and 4.0*10–4, respectively. The brain phantom reconstructions with multi-pinhole collimator are superior as compared to slit-slat, especially in terms of symmetry and realistic activity distribution. DOI information reduces artefacts and improves uniformity in geometric phantoms, although the difference is small for the brain phantom. These results favour a multi-pinhole configuration. Figure 1 Top left: trans-axial sections of a reconstructed uniform phantom. Top right: sagittal sections of a reconstructed Defrise phantom. Bottom left: transaxial sections of a reconstructed Derenzo phantom. Bottom right: transaxial (left), coronal (middle) ...

On Artifact-Free Projection Overlaps in Multi-Pinhole Tomographic Imaging

In multi-pinhole SPECT, overlapping the projections from different pinholes has been used to increase sensitivity. However, the prevailing view is that the overall quality of the reconstructed image is not improved by the overlaps in the projections. It is often stated in literatures that overlaps introduce ambiguous information, which can lead to nonuniqueness of solution for the inverse problem, and thus artifacts are introduced in the reconstructed image. On the other hand, contrary to the prevailing view, a recent study on slit-slat collimators shows that artifacts can be removed with the "help" of an extra complete nonoverlapped projection data set. In this paper, two types of artifact-free projection overlaps are defined in general, and the criteria for designing artifact-free multi-pinhole systems with overlaps are proposed. It is shown that once the criteria are satisfied, the solution of the inverse problem is unique, and thus no artifact is expected in the reconstructed image. Via extensive simulation study, various artifact-free overlapping multi-pinhole systems are designed and validated. It is shown that overlaps in the artifact-free systems can improve contrast-to-noise ratio (CNR). With a proper design, the CNR for an artifact-free overlapping system can be significantly higher than that for the corresponding nonoverlapping system. The improved image quality is also confirmed with noisy reconstructions.

Slit-Slat Collimator Equipped Gamma Camera for Whole-Mouse SPECT-CT Imaging

A slit-slat collimator was developed for a gamma camera intended for small animal imaging (mice). The gamma camera contains a 10 cm × 10 cm pixelated NaI(Tl) crystal array of individual 1.3 mm × 1.3 mm × 6 mm crystal elements, distanced at 0.2 mm. The housing of the collimator was made of 5 mm tungsten alloy, forming a slit with a 1 mm opening and a 90 degree acceptance angle. The slats were made of lead foils with 0.2 mm in thickness to match the crystal spacing, and possess a length of 40 mm. Individual slats are separated by polyurethane material with a thickness of 1.3 mm corresponding to the size of the crystal elements. Alignment of the collimator with the camera's crystal pattern was performed by adjusting a micrometer screw while monitoring a Co-57 point source for maximum signal intensity. For SPECT, the collimator forms a cylindrical FOV enabling whole mouse imaging with a magnification of 1.9 at COR over the entire AOR (magnification = 1.0). As the gamma camera is part of a bimodal imaging system incorporating also x-ray CT capabilities the collimator had to be co-registered with the other components to yield intrinsically co-registered reconstruction results of both modalities. For this, five parameters corresponding to the geometrical displacements of the collimator as well as the mechanical co-alignment between the gamma camera and the CT subsystem were estimated by means of a dedicated measurement setup involving a number of bimodal calibration sources. To illustrate the performance of the slit-slat collimator and compare its performance to a single pinhole collimator a Derenzo phantom study has been performed. Transversal resolution along the entire long axis is comparable to a pinhole collimator of same pinhole diameter. Axial resolution of the slit-slat collimator is comparable to that of a parallel beam collimator. In addition, a mice study has been performed, as well, with data being presented.

Finding Optimized Conditions of Slit-Slat and Multislit-Slat Collimation for Breast Imaging

In order to develop a breast-imaging system for Single Photon Emission Computed Tomography (SPECT) using slit-slat and multislit-slat collimators, we searched for optimized geometric parameters of the collimators. For this study, we employed two independent metrics to validate each result: 1) Signal to Noise Ratio (SNR) based on the Cramer-Rao lower Bound (CRB) and 2) contrast at the same noise level from an ensemble. We calculated SNR values using forward-projection data of an anthropomorphic digital phantom containing two lesions in the breast (one at the chest wall and the other at the center) with a simulated slit-slat collimator as a function of the collimator's geometric parameters. We also calculated contrast values from reconstructed images with noise. Based on the results from the slit-slat case, we investigated angular range, SNR, and contrast for the multislit-slat. We saw similar trends of the two metrics. One interesting property of the multislit-slat is that the imaging performance depends on the orientation of the field of view (FOV) of the side slits. When we compared the metric values for the slit-slat and multislit-slat, improvement was seen only when the lesion was in the FOV of the side slits. Therefore, tuning the parameters of the multislit-slat to optimally detect lesions at the chest wall might be a sensible option since the slit-slat already provides good image quality for center and superficial lesions.

Optimization and Calibration of Slat Position for a SPECT With Slit-Slat Collimator and Pixelated Detector Crystals

To expand the availability of SPECT for biomedical research, we developed a SPECT imaging system on an existing animal PET detector by adding a slit-slat collimator. As the detector crystals are pixelated, the relative slat-to-crystal position (SCP) in the axial direction affects the photon flux distribution onto the crystals. The accurate knowledge of SCP is important to the axial resolution and sensitivity of the system. This work presents a method for optimizing SCP in system design and for determining SCP in system geometrical calibration. The optimization was achieved by finding the SCP that provides higher spatial resolution in terms of average-root-mean-square ( <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R̅M̅S̅</i> ) width of the axial point spread function (PSF) without loss of sensitivity. The calibration was based on the least-square-error method that minimizes the difference between the measured and modeled axial point spread projections. The uniqueness and accuracy of the calibration results were validated through a singular value decomposition (SVD) based approach. Both the optimization and calibration techniques were evaluated with Monte Carlo (MC) simulated data. We showed that the [R̅M̅S̅] was improved about 15% with the optimal SCP as compared to the least-optimal SCP, and system sensitivity was not affected by SCP. The SCP error achieved by the proposed calibration method was less than 0.04 mm. The calibrated SCP value was used in MC simulation to generate the system matrix which was used for image reconstruction. The images of simulated phantoms showed the expected resolution performance and were artifact free. We conclude that the proposed optimization and calibration method is effective for the slit-slat collimator based SPECT systems.

Experimental results from a prototype slit-slat collimator with mixed multiplexed and non-multiplexed data

We have previously shown with simulations that a gain in signal-to-noise ratio (SNR) can be obtained by using mixed multiplexed (MX) and non-MX data in a slit-slat SPECT system as compared to using non-MX data only. We have now developed a prototype slit-slat collimator for a conventional gamma camera in order to validate these simulation results. The prototype collimator consists of seven slits and multiple parallel slats. Image reconstruction is performed using a modified OSEM algorithm, which takes into account geometric sensitivity variations and attenuation, but not scatter or resolution effects. Here, we first describe the calibration of the system and then we present the experimental validation with phantom experiments. SPECT acquisitions using different geometric and anthropomorphic phantoms were performed with and without multiplexing. The results show that reconstruction of the MX projections with the non-MX-projections eliminates artefacts caused by multiplexing. SNR gains obtained using the mixed MX and non-MX configurations were in the range of 26% to 51% for different phantoms. The results were in agreement with our previously published simulation work, proving that combining MX and non-MX data can result in artefact-free reconstructed images with improved SNR.

SU-E-I-182: Slit-Slat Collimators in a Staggered Array for Dedicated Breast SPECT

Purpose: To compare slit‐slat collimators in single‐layer and staggered arrays with standard parallel‐beam collimators in a dedicated breast SPECT system. Methods: A ring of slit‐slat collimators surrounds a cylindrical region of 20 cm diameter for SPECTimaging of a pendant breast. A second ring of cameras, viewing between cameras in the first ring, may be added to give a staggered array of cameras. Such arrays could be static or rotating; for now, we assume a rotating array, for a radially symmetric ray pattern. Slit width and slat design are specified to match the spatial resolution of a low‐energy all‐purpose parallel‐hole (LEAP) collimator at the axis of rotation. Sensitivity is calculated for arrays of slit‐slat and parallel‐hole collimated cameras, using the maximum number of cameras possible for each. Results: With focal length of 5.7 cm, detector width of 6.5 cm, and 15 cm radius of rotation (ROR), the slit‐slat array could contain twenty cameras for one ring, and forty cameras for two rings. A LEAP‐collimated system with 10 cm ROR could contain four cameras and would have a spatial resolution of 8.0 mm at the axis of rotation. To match this, the slit‐slat collimators would have slit widths of 2.2mm/1.8mm for first and second rows respectively, and slats designed for similar axial resolution. A single row of twenty slit‐slat collimated cameras would have a sensitivity of 8.66×10, −4photons transmitted per photon emitted, while a staggered array would have a sensitivity of 1.28×10, −3. The array of four LEAP‐collimated cameras would have a sensitivity of 6.48 ×10, −4. Conclusions: The single‐ ring configuration of cameras gives sensitivity 34% greater than the LEAP‐ collimated system, while the staggered array gives a 98% sensitivity increase. This doubling in sensitivity comes with a large increase in complexity. Further optimizations for the slit‐slat system are possible in both areas.

An elliptical SPECT system with slit–slat collimation for cardiac imaging

Cardiac studies are a good candidate for SPECT (single photon emission computed tomography) because of the large clinical demand and the need for improved image quality. But SPECT imaging suffers from poor spatial resolution and high statistical noise. A new SPECT system with slit–slat collimation arranged on an elliptical arc for cardiac imaging is proposed in this paper. Simulated emission computed tomography data are generated along an elliptical moving orbit with system configuration parameters. The iterative reconstruction techniques are used to implement the cardiac imaging of the proposed SPECT system. Image reconstruction can be done using the OS-EM algorithm from the data collected. This system is developed to improve the reconstructed image if attenuation correction and depth-dependent correction are included in the reconstruction, and the body contour is used in the reconstruction for severely truncated data. The simulation results show that the present methods can provide significant improvement in the spatial resolution and the image quality of SPECT sets.

Slit-Slat and Multi-Slit-Slat Collimator Design and Experimentally Acquired Phantom Images From a Rotating Prototype

We have previously found and validated expressions for slit-slat (SS) geometric efficiency and resolution. These expressions have suggested that SS may be a good choice for imaging mid-size objects or objects that are long axially since (i) the geometric efficiency increases near the slit as h(-1) (instead of h(-2) for pinhole (PIN) and either decreases near the collimator for fan-beam (FB) or remains constant for parallel-beam (PB)), where h is the distance from a point to the slit plane; (ii) the transverse resolution is comparable to that of PIN, which is better than that of FB and PB for small objects; (iii) the axial resolution is worse than that of PIN since there is no axial magnification; (iv) there is a large axial FOV, unlike PIN, which is likely to be useful when imaging mid-size or long objects; and (v) there is no need for 3D orbits (e.g., helical) since each slice is complete (like PB and FB).We have developed a rotating prototype SS collimator that is capable of single-slit or multi-slit acquisition of data. The focal length (FL) is shorter than that of a typical PIN since increasing the FL requires taller slats to maintain resolution; taller slats reduce geometric efficiency. A lead rectangular box was used to provide support and shielding around the slit-slat collimator. Lead slats, spaced with Rohacell foam, were mounted in an assembly with 3 mm pitch.We have performed preliminary characterization with point sources and acquired micro hot- and cold-rod phantoms and a Deluxe Jaszczak phantom. The projections have been reconstructed using an MLEM algorithm and show good resolution.Comparisons indicate that SS is more sensitive than PB and FB for the same resolution for smaller-diameter objects. The advantage of SS over PB and FB increases as the desired resolution improves. SS can also be used in configurations that yield projections that have non-isotropic resolution; it is possible for SS to achieve transverse resolutions that are unreachable by PB, since PB does not magnify, and by FB, since its magnification factor for small objects is much smaller than that of SS. Experimental results show that the resolution of the reconstructed phantoms is consistent with theoretical expectations.

Derivation of System Matrix From Simulation Data for an Animal SPECT With Slit-Slat Collimator

We developed SPECT imaging capability on an animal PET system. Our goal was to provide animal PET users the SPECT capability at a low cost and facilitate potential PET/SPECT dual modality imaging applications. The SPECT function was enabled with a slit-slat collimator insert and by acquiring data in singles mode. The focus of this paper is to establish a method for deriving the system matrix for the SPECT system from Monte Carlo simulation. With the Monte Carlo package GATE, we simulated a uniform cylinder source which filled the SPECT field of view (FOV). To reduce the size of the original large and sparse system matrix, the detectors that were exposed to individual emission elements were selectively included for system matrix derivation and storage. The axial symmetry of the system was exploited so that only the base-axial volume was used for deriving system response. The system matrix derived was validated with point source measurements at known positions and implemented in an iterative reconstruction algorithm. The imaging performance of the system matrix was evaluated with experimental phantom studies. Reconstructed phantom images were artifact free and demonstrated expected spatial resolution. The method presented in this work is generally applicable to other SPECT imaging systems.

A SVD-Based Method to Assess the Uniqueness and Accuracy of SPECT Geometrical Calibration

Geometrical calibration is critical to obtaining high resolution and artifact-free reconstructed image for SPECT and CT systems. Most published calibration methods use analytical approach to determine the uniqueness condition for a specific calibration problem, and the calibration accuracy is often evaluated through empirical studies. In this work, we present a general method to assess the characteristics of both the uniqueness and the quantitative accuracy of the calibration. The method uses a singular value decomposition (SVD) based approach to analyze the Jacobian matrix from a least-square cost function for the calibration. With this method, the uniqueness of the calibration can be identified by assessing the nonsingularity of the Jacobian matrix, and the estimation accuracy of the calibration parameters can be quantified by analyzing the SVD components. A direct application of this method is that the efficacy of a calibration configuration can be quantitatively evaluated by choosing a figure-of-merit, e.g., the minimum required number of projection samplings to achieve desired calibration accuracy. The proposed method was validated with a slit-slat SPECT system through numerical simulation studies and experimental measurements with point sources and an ultra-micro hot-rod phantom. The predicted calibration accuracy from the numerical studies was confirmed by the experimental point source calibrations at approximately 0.1 mm for both the center of rotation (COR) estimation of a rotation stage and the slit aperture position (SAP) estimation of a slit-slat collimator by an optimized system calibration protocol. The reconstructed images of a hot rod phantom showed satisfactory spatial resolution with a proper calibration and showed visible resolution degradation with artificially introduced 0.3 mm COR estimation error. The proposed method can be applied to other SPECT and CT imaging systems to analyze calibration method assessment and calibration protocol optimization.

Comparison of Circular and Polygonal Planar Orbits for Pinhole and Slit-Slat SPECT

Analytic formulas and Monte Carlo simulation are used to calculate and compare the sensitivity of circular and polygonal orbits at different points in the Field of View (FoV) for both pinhole and slit-slat collimation. Results show that for a given slit-slat collimator an N-sided polygonal orbit tangent to the FoV generally provides average sensitivity lower than the tightest circular orbit consistent with the same aperture angle, but with better spatial resolution that can be traded for sensitivity for a constant-resolution comparison. This generally results in a slight advantage for the polygonal orbit. However, this advantage depends on the clearance that must be allowed between the orbit and the FoV and decreases quickly, vanishing when even a few millimeters of space are left, which in practice is necessary to accommodate mechanical constraints. For a pinhole collimator the advantage for the tangent polygonal orbit is more consistent, but similar conclusions are reached again when clearance is considered. A direct comparison at constant resolution between slit-slat and pinhole collimation in a single transverse plane is shown to be possible with parameters typical of small-animal imaging applications only for detectors with excellent intrinsic resolution; in this case, pinhole collimation is shown to be more sensitive in magnifying geometries, but reduced axial FoV and increased axial blurring should also be considered for a more complete comparison.