Search

Background Purpose Methods Results Conclusion Photon counting detectors (PCDs) for x‐ray computed tomography (CT) are the future of CT imaging. At present, semiconductor‐based PCDs such as cadmium telluride (CdTe), cadmium zinc telluride, and silicon have been either used or investigated for clinical PCD CT. Unfortunately, all of them have the same major challenges, namely high cost and limited spectral signal‐to‐noise ratio (SNR). Recent studies showed that some high‐quality scintillators, such as lanthanum bromide doped with cerium (LaBr3:Ce), are less expensive and almost as fast as CdTe.The objective of this study is to assess the performance of a LaBr3:Ce PCD for clinical x‐ray CT.We performed Monte Carlo simulations and compared the performance of 3 mm thick LaBr3:Ce and 2 mm thick CdTe for PCD CT with x‐rays at 120 kVp and 20–1000 mA. The two PCDs were operated with either a threshold–subtract (TS) counting scheme or a direct energy binning (DB) counting scheme. The performance was assessed in terms of the accuracy of registered spectra, counting capability, and count‐rate‐dependent spectral imaging‐task performance, for conventional CT imaging, water–bone material decomposition, and K‐edge imaging with tungsten as the K‐edge material. The performance for these imaging‐tasks was quantified by nCRLB, that is, the Cramér–Rao lower bound on the variance of basis line‐integral estimation, normalized by the corresponding value of CdTe at 20 mA.The spectrum recorded by CdTe was distorted significantly due to charge sharing, whereas the spectra recorded by LaBr3:Ce better matched the incident spectrum. The dead time, estimated by fitting a paralyzable detector model to the count‐rate curves, was 20.7, 15.0, 37.2, and 13.0 ns for CdTe with TS, CdTe with DB, LaBr3:Ce with TS, and LaBr3:Ce with DB, respectively. Conventional CT imaging showed an adverse effect of reduced geometrical efficiency due to optical reflectors in LaBr3:Ce PCD. The nCRLBs (a lower value indicates a better SNR) for CdTe with TS, CdTe with DB, LaBr3:Ce with TS, LaBr3:Ce with DB, and the ideal PCD, were 1.00 ± 0.01, 1.00 ± 0.01, 1.18 ± 0.02, 1.18 ± 0.02, and 0.79 ± 0.01, respectively, at 20 mA. The nCRLBs for water–bone material decomposition, in the same order, were 1.00 ± 0.02, 1.00 ± 0.02, 0.85 ± 0.02, 0.85 ± 0.02, and 0.24 ± 0.02, respectively, at 20 mA; and 0.98 ± 0.02, 0.98 ± 0.02, 1.09 ± 0.02, 0.83 ± 0.02, and 0.24 ± 0.02, respectively, at 1000 mA. Finally, the nCRLBs for K‐edge imaging, the most demanding task among the five, were 1.00 ± 0.02, 1.00 ± 0.02, 0.55 ± 0.02, 0.55 ± 0.02, and 0.13 ± 0.02, respectively, at 20 mA; and 2.45 ± 0.02, 2.29 ± 0.02, 3.12 ± 0.02, 2.11 ± 0.02, and 0.13 ± 0.02, respectively, at 1,000 mA.The Monte Carlo simulations showed that, compared to CdTe with either TS or DB, LaBr3:Ce with DB provided more accurate spectra, comparable or better counting capability, and superior spectral imaging‐task performances, that is, water–bone material decomposition and K‐edge imaging. CdTe had a better performance than LaBr3:Ce for the conventional CT imaging task due to its higher geometrical efficiency. LaBr3:Ce PCD with DB scheme may be an excellent alternative option for CdTe PCD. [ABSTRACT FROM AUTHOR]

A Compton camera prototype for ion beam range monitoring via prompt (< 1 ns) gamma detection in hadron therapy is being developed and characterized at the Medical Physics Department of LMU Munich. The system consists of a large (50x50x30 mm3) monolithic LaBr3(Ce) scintillation crystal as absorber component to detect the multi-MeV Compton scattered photons, together with a stack of 6 double-sided silicon strip detectors (DSSSD) acting as scatterer component. Key ingredient of the γ-source reconstruction is the determination of the γ-ray interaction position in the scintillator, which is read out by a 256-fold segmented multi-anode photomultiplier tube (PMT). From simulations an angular resolution of about 1.5o for the photon source reconstruction can be expected for the energy range around 3 – 5 MeV, provided that a spatial resolution of 3 mm can be reached in the absorbing scintillator [1]. Therefore, particular effort was dedicated to characterize this latter property as a function of the γ-ray energy. Intense, tightly collimated 137Cs and 60Co photon sources were used for 2D irradiation scans (step size 0.5 mm) as prerequisite for studying the performance of the “k-Nearest-Neighbors” algorithm developed at TU Delft [2] (together with its variant ”Categorical Average Pattern”, CAP) and extending its applicability into the energy range beyond the original 511 keV. In this paper we present our most recent interaction position analysis in the absorbing scintillator, leading to a considerably improved value for the spatial resolution: systematic studies were performed as a function of the k-NN parameters and the PMT segmentation. A trend of improving spatial resolution with increasing photon energy was confirmed, resulting in the realization of the presently optimum spatial resolution of 2.9(1) mm @1.3 MeV, thus reaching the design specifications of the Compton camera absorber. The specification goal was reached also for a reduced PMT segmentation of 8x8 anode segments (each with 6x6 mm2 active area), thus allowing to reduce the complexity of the signal processing while preserving the performance.

Neural Networks trained with error back propagation Levenberg-Marquardt training (LM), Neural Networks trained with an algebraic method and Support Vector Machines (SVM) were evaluated to extract the position information from measured light distributions generated by the interactions of 511 keV photons in monolithic scintillator blocks. All three algorithms can achieve a similar average resolution (~1.6 mm FWHM in a 20 x 10 x 10 mm LSO block) but the LM trained neural networks do so most efficiently. When the incidence angle of the photons increases to 30[degrees], the resolution degrades slightly to 2.0 mm FWHM. A small mismatch (< [+ or -] 5[degrees]) between the true incidence angle and the angle for which a neural network was trained can be tolerated without significant resolution loss. Increasing the thickness to 20 mm and using a top-bottom readout of the block yields an average resolution of 2.2 nun FWHM. Index Terms--Monolithic scintillator, positron emission tomography, neural network.

Monolithic scintillator detectors, consisting of several [cm.sup.3] of scintillating material coupled to one or more Hamamatsu S8550 avalanche photodiode (APD) arrays, are proposed as detectors for high resolution positron emission tomography (PET). In this work, the factors contributing to the variance on the signals are investigated, and their effects on the energy, time and spatial resolutions are analyzed. Good agreement was found between a model of the energy resolution and experiments with a 20 x 10 x 10 [mm.sup.3] LYSO:Ce crystal coupled to a single channel large-area AID (LAAPD). With the same crystal coupled to an APD array, differences between model and experiment were observed at high APD gain. The measured energy resolution of ~11% FWHM was dominated by scintillation photon statistics, with less important roles for the APD excess noise factor and electronic noise. On the other hand, electronic noise was an important factor both for the time and the spatial resolutions. The time resolution was found to depend strongly on the APD bias voltage, and was best at the highest bias. A time resolution of 1.6 ns full width at haft maximum (FWHM) was measured against a [BaF.sub.2]-PMT detector. The best spatial resolution measured was 1.64 mm FWHM, without correction for the ~0.9 mm FWHM measurement beam. It is estimated that an intrinsic spatial resolution of 1.26 mm FWHM can be achieved at the center of the detector with an infinitely narrow test beam. Index Terms--Avalanche photodiode (APD), monolithic scintillator detector, positron emission tomography (PET), signal to noise ratio (SNR).

Previous experimental results have demonstrated that a PET detector module based on a 20 X 10 x 10 nun LSO block read out by a Hamamatsu $8550 APD array is able to determine the impinging position of perpendicularly incident 511--keV photons with a resolution better than 2-mm FWHM. A prototype PET demonstrator using two of these detector heads was build and evaluated. The axial and transaxial resolution measured in two-dimensional (2-D) OSEM and 2-D FBP reconstructed images is better than 1.5- and 2-mm FWHM, respectively, for radial distances less than 15 mm. A Monte Carlo simulation showed that a four-ring PET scanner based on these monolithic scintillator blocks has a sensitivity which is about 2.5 times higher than the sensitivity of a similar system equipped with detector heads using a 4 x 8 matrix of 2 x 2 X 10 mm LSO pixels. Index Terms--Light sharing, neural network, PET, scintillator.