IEEE Nuclear Science Symposium conference record. Nuclear Science Symposium最新文献

Our current approach to fusion of CTCA and PET perfusion data uses the epicardial surface from the perfusion data onto which the CT coronary arteries are aligned and warped. This work was undertaken to improve the alignment and the display realism by using CT epicardial boundary information. PET and CTCA images from a combined scanner were used. Based on the location of the LV detected from PET during standard perfusion processing, the LV chamber of the CT was located. Hounsfield units were used to define an estimated endocardial surface in the CT. Based on the endocardial surface, the epicardial boundary was detected, again using Hounsfield units, or when that failed, by estimating its position based on the detected endocardium. Coronary arteries were detected using a commercial program; the epicardial surface was forced to be congruent with all detected artery points. A confidence factor in each epicardial boundary point was maintained based on how each was detected, whether through threshold, through estimation, or by using he coronary artery points. The epicardial boundary surface points were nonlinearly filtered; erroneous surface points, as defined by local properties and confidence factors, were replaced with values interpolated from the nearest points deemed more accurate. The resulting epicardial surface was linearly aligned to the epicardial boundary detected from the PET, and the CT boundaries were then color-coded based on the PET perfusion. Resulting surfaces were much more realistic than those created using PET epicardial boundaries (Fig 1.) Forcing the CT epicardial surface to lie on the detected coronary arteries eliminated problems with alignment and warping of the coronary arteries onto the PET surface.

The objective of this investigation is to determine the impact of respiratory motion on the detection of small solitary pulmonary nodules (SPN) in single photon emission computed tomographic (SPECT) imaging. We have previously modeled the respiratory motion of SPN based on the change of location of anatomic structures within the lungs identified on breath-held CT images of volunteers acquired at two different stages of respiration. This information on respiratory motion within the lungs was combined with the end-expiration and time-averaged NCAT phantoms to allow the creation of source and attenuation maps for the normal background distribution of Tc-99m NeoTect. With the source and attenuation distribution thus defined, the SIMIND Monte Carlo program was used to produce SPECT projection data for the normal background and separately for each of 150 end-expiration and time-averaged simulated 1.0 cm tumors. Normal and tumor SPECT projection sets each containing one lesion were combined with a clinically realistic noise level and counts. These were reconstructed with RBI-EM using 1) no correction (NC), 2) attenuation correction (AC), 3) detector response correction (RC), and 4) attenuation correction, detector response correction, and scatter correction (AC_RC_SC). The post-reconstruction parameters of number of iterations and 3-D Gaussian filtering were optimized by human-observer studies. Comparison of lesion detection by human-observer LROC studies reveals that respiratory motion degrades tumor detection for all four reconstruction strategies, and that the magnitude of this effect is greatest for NC and RC, and least for AC_RC_SC. Additionally, the AC_RC_SC strategy results in the best detection of lesions.

Our laboratory has been developing a depth-of-interaction (DOI) detector design based on light sharing between pairs or quadlets of crystals. Work to date has been utilizing 2×2 mm cross section crystals carefully positioned on a multi-anode PMT. However, there is still significant light sharing in the PMT glass envelope and current PMT designs do not allow one-on-one coupling for arrays of smaller cross section crystals. One-on-one coupling is optimal for implementing the DOI estimator. An alternative to PMTs is to take advantage of progress in fabrication of metal resistive-layer semiconductor photodetectors to provide arrays with one-on-one crystal coupling. We report on our initial tests of one manufacturer's devices. The photodetector (MAPD) and scintillator combination (LFS-3) are both products of Zecotek. The LFS-3 crystal is a variant of LFS that has a better spectral match to the MAPD. Measurements show performance equivalent to or better than that obtained with PMTs and LSO, LFS, or LYSO crystals. For example, 2×2×20 mm crystals are providing 11% energy resolution. The high gain of such devices allow flexibility in designs for both the array and the supporting electronics. We are proceeding with the dMiCE development based on the use of MAPD and LFS-3 arrays.

Expanding on the work of Nuyts et. al [1], Bai et. al. [2], and Bai and Shao [3], who all studied the effects of attenuation and attenuation correction on tumor-to-background ratios and signal detection, we have derived a general expression for the tumor-to-background ratio (TBR) for SPECT attenuated data that have been reconstructed with a linear, non-iterative reconstruction operator O. A special case of this is when O represents discrete filtered back-projection (FBP). The TBR of the reconstructed, uncorrected attenuated data (TBR(no-AC)) can be written as a weighted sum of the TBR of the FBP-reconstructed unattenuated data (TBR(FBP)) and the TBR of the FBP-reconstructed "difference" projection data (TBR(diff)). We evaluated the expression for TBR(no-AC) for a variety of objects and attenuation conditions. The ideal observer signal-to-noise ratio (SNR(ideal)) was also computed in projection space, in order to obtain an upper bound on signal detectability for a signal-known-exactly/background-known-exactly (SKE/BKE) detection task. The results generally show that SNR(ideal) is lower for tumors located deeper within the attenuating medium and increases for tumors nearer the edge of the object. In addition, larger values for the uniform attenuation coefficient μ lead to lower values for SNR(ideal). The TBR for FBP-reconstructed, uncorrected attenuated data can both under- and over-estimate the true TBR, depending on several properties of the attenuating medium, including the shape of the attenuator, the uniformity of the attenuator, and the degree to which the data are attenuated.

We present the derivation of a live-time model for predicting count rates in computer simulations of PET scanners. Computer models are frequently used to investigate new PET scanner configurations, but they typically do not account for the count losses caused by scanner-specific electronics and processing. The live-time fraction depends strongly on the photon flux incident on the detector. We modeled the live-time of a clinical PET scanner by relating measured and simulated single photon fluxes. Our model used data from a specific scanner, but the approach is generally applicable.We applied the live-time model to partial collimation on a PET scanner; in particular, a scanner with septa positioned between every third detector ring ("2.7D" acquisition mode). The photon flux was measured and simulated for conventional acquisition modes (2D, 3D), and simulated for partial collimation (2.7D). These data were used in the model to predict live-time for partial collimation. The model was then validated against measurements in 2.7D mode. At low activity the model was very accurate at predicting the live-time fraction. Over-estimation of count-rates by the simulations lead to an uncertainly in the live-model. The uncertainty increased with activity concentration, reaching 0.9% and 2.2% at 20 kBq/mL for singles and coincidence live-time, respectively. When applied to 2.7D mode, the model predicted coincidence live-time accurate to 2.2% and 10% at 5 kBq/mL and 20 kBq/mL in the phantom, respectively. The 2.7D singles-counting live-time was predicted to within 0.2% of the measured value for up to 20 kBq/mL in the phantom.

The problem we address is the optimization and comparison of window-based scatter correction (SC) methods in SPECT for maximum a posteriori reconstructions. While sophisticated reconstruction-based SC methods are available, the commonly used window-based SC methods are fast, easy to use, and perform reasonably well. Rather than subtracting a scatter estimate from the measured sinogram and then reconstructing, we use an ensemble approach and model the mean scatter sinogram in the likelihood function. This mean scatter sinogram estimate, computed from satellite window data, is itself inexact (noisy). Therefore two sources of noise, that due to Poisson noise of unscattered photons and that due to the model error in the scatter estimate, are propagated into the reconstruction. The optimization and comparison is driven by a figure of merit, the area under the LROC curve (ALROC) that gauges performance in a signal detection plus localization task. We use model observers to perform the task. This usually entails laborious generation of many sample reconstructions, but in this work, we instead develop a theoretical approach that allows one to rapidly compute ALROC given known information about the imaging system and the scatter correction scheme. A critical step in the theory approach is to predict additional (above that due to to the propagated Poisson noise of the primary photons) contributions to the reconstructed image covariance due to scatter (model error) noise. Simulations show that our theory method yields, for a range of search tolerances, LROC curves and ALROC values in close agreement to that obtained using model observer responses obtained from sample reconstruction methods. This opens the door to rapid comparison of different window-based SC methods and to optimizing the parameters (including window placement and size, scatter sinogram smoothing kernel) of the SC method.

Hybrid LROC studies can be used to more realistically assess the impact of reconstruction strategies, compared to those constructed with digital phantoms. This is because hybrid data provides the background variability that is present in clinical imaging, as well as, control over critical imaging parameters, required to conduct meaningful tests. Hybrid data is obtained by adding Monte Carlo simulated lesions to disease free clinical projection data. Due to Ga-67 being a particularly challenging radionuclide for imaging, we use Ga-67 hybrid SPECT data to study the effectiveness of the various correction strategies developed to account for degradations in SPECT imaging. Our data was obtained using GE-VG dual detector SPECT-CT camera. After determining a target lesion contrast we conduct pilot LROC studies to obtain a near-optimal set of reconstruction parameters for the different strategies individually. These near-optimal parameters are then used to reconstruct the final evaluation study sets. All LROC study results reported here were obtained employing human observers only. We use final LROC study results to assess the impact of attenuation compensation, scatter compensation and detector resolution compensation on data reconstructed with the RBI-EM algorithm. We also compare these with FBP reconstructions of the same dataset. Our experiment indicates an improvement in detection accuracy, as various degradations inherent in the image acquisition process are compensated for in the reconstruction process.

We present a study that introduces two approaches to implementing block detectors into SimSET and compares their performance. SimSET is a photon tracking simulation package, which currently incorporates only detectors made of a solid annulus of scinitillator material. A pseudo-block approximation has been imposed on the solid annulus of conventional SimSET by discarding interactions in annulus segments that span the angular block gap. This yields blocks that are annulus segments, not rectangles. This is a quick and easy approximation of block structure, which brings SimSET results closer to actual scanner measurements. Even better agreement is expected with a deeper modification of the SimSET code that implements true rectangular blocks in the detector module (to be released late 2007/early 2008). This approach enables the greatest amount of variability and trueness to detail.We compare results from both block structure implementations to the conventional SimSET results and to measurements from a GE DSTE PET/CT scanner. Differences are evaluated in terms of sensitivities, crystal maps, and energy spectra, as well as in benchmark time tests of the simulation runs and their ease of use.Either implementation of block structure can aid in improving simulation accuracy by ameliorating one known cause of discrepancies, the geometric nature of the block detectors.

Respiratory motion in PET/CT imaging degrades PET image quantitation due to misaligned attenuation correction (AC) factors and motion blurring. This work explores the use of the Radon consistency conditions to compensate for these limitations in respiratory gated PET images in which only a single CT scan is available for AC. Specifically, we use the Radon consistency of AC-PET data as a metric to transform the attenuation map to match each phase of respiratory gated data, perform phase matched AC, and then use the inverse of the transformation parameters to align the gated PET images into a single phase. A final image volume is formed from summing PET images aligned to a single phase. We test this method with three transformation types applied to simulated data and measured patient PET/CT data. Results show successful alignment of attenuation maps and minor quantitative improvement with the proposed methods.

The quality of SPECT images suffers from the effects of photon attenuation and scatter, and from distance-dependent collimator blur, and many researchers have shown the benefit of compensating for these degradations in the inverse problem. For this work, we examined how using an incorrect collimator-blur model affects the detection and localization of (67)Ga-avid lymphomas in simulated chest scans. In particular, we considered whether blur-overcompensation can enhance reconstructed images for purposes of localizing tumors. Variations in the correct linear model for medium-energy, parallel-hole collimators were compared by means of LROC studies with human and localizing model observers. Imaging data consisted of Simind projections of the MCAT phantom, and RBI reconstructions were performed. Our results indicate that tumor-detection performance is not improved by using a mismatched RC model. Reconstruction with increased RC requires more iterations, which leads to longer noise correlations. Our results also suggest a substantial observer insensitivity to the accuracy of the RC model.