Jinhun Joung

Implementation of ML based positioning algorithms for scintillation cameras

Positioning algorithms for scintillation cameras are investigated. Maximum Likelihood (ML) estimation techniques and centroid-based methods (e.g., weighted and local centroid) are evaluated. Three implementation methods to reach the two-dimensional ML solution are proposed. First, the authors investigate a 1D based recursive ML positioning algorithm. The technique overcomes the dimensional separable violation for hexagonally packed photomultiplier tube (PMT) arrays. Second, they examine a 2D based ML method that uses local PMT clusters to reduce the computational demands of full 2D ML implementation. Third, they develop a correlation method that maps event characterization vectors to the associated position. The methods are tested with data sets generated using Monte Carlo simulation and verified with experimental study. Simulation (experimental) results show that the full 2D ML algorithm is superior to the others. It has a 7(10)%, 67(47)% and 69(30)% improvement over the weighted centroid with 3% bias subtraction method in terms of spatial resolution, linearity and MSE, respectively.

Investigation of bias-free positioning estimators for the scintillation cameras

Bias-free positioning estimators for scintillation cameras are investigated. A linear correlation coefficient (LCC) and Chi square error (CSE) method are evaluated with respect to linearity and spatial resolution performance. The LCC method uses correlation information between the true function (i.e., light response function) and measured data in mapping an event characterization vector to the associated position. The CSE method estimates the position where the Chi square error between two functions becomes minimized. In order to determine true statistics as a function of position, the light response function (LRF) was estimated based on sample measurements by using a cubic spline interpolation technique which provides smooth first-order and continuous second-order derivative of the LRF. Both methods have superior linearity properties compared to the weighted centroid. Each method call be considered as a bias-free positioning estimator within the effective field of view of the detector. The spatial resolution performance of the CSE method is /spl sim/7% and /spl sim/16% better than the weighted centroid method for a 16- and 25-mm-thick crystal, respectively. The spatial resolution performance of the LCC method is comparable to that of the weighted centroid method.

cMiCE:a high resolution animal PET using continuous LSO with a statistics based positioning scheme

We have investigated the feasibility of using a continuous miniature crystal element (cMiCE) detector module for high-resolution small animal PET applications. The detector module consists of a single continuous slab of LSO, 25/spl times/25 mm in exposed cross-section and 4 mm thick, coupled directly to a PS-PMT (Hamamatsu R5900-00-C12). Further, a new statistics based positioning (SBP) algorithm has been implemented to address linearity and edge effect artifacts that are inherent with conventional Anger style positioning schemes. Detector performance has been evaluated for both SBP and Anger based decoding using measured data and Monte Carlo simulations. Using the SBP scheme, edge artifacts were successfully handled. Simulation results show that the useful field of view (UFOV) was extended to /spl sim/22/spl times/22 mm with an average point spread function of /spl sim/0.5 mm full width at half maximum (FWHM/sub PSF/). For the same detector with Anger decoding the UFOV of the detector was /spl sim/16/spl times/16 mm with an average FWRM/sub PSP/ of 4.9 mm. Experimental results yielded similar differences between FOV and resolution performance. FWHM/sub PSF/ for the SBP and Anger based method was 1.4 mm and 2.0 mm, uncorrected for source size, with a 1 mm diameter point source, respectively. The results demonstrate that the SBP scheme yields improved performance over traditional Anger techniques for our cMiCE detector.

Performance characteristics of a second generation micro crystal element (MiCE2) detector

This work reports on performance characteristics of a second generation micro crystal element (MiCE2) detector for a dedicated PET system to image mice. Our MiCE2 detector consists of a 22/spl times/22, array of 0.8/spl times/0.8/spl times/6 mm mixed lutetium silicate (MLS) crystals. Five sides of the crystals are polished with one 0.80.8 mm face left unpolished. The crystals are placed within a grid made of a highly reflective polymer film material. The grid optically isolates the crystals and also functions as a reflective wrap. The detector unit is directly coupled to a 6+6 cross-anode position sensitive PMT. The crystal of interaction is determined using simple Anger style logic. Crystal maps have been created for a 22/spl times/22 crystal array. All 484 crystals are visualized in the full detector modules crystal map. The average peak to valley ratio between neighboring crystals was 6.4. There was greater than a factor of two difference between the photopeak energy channel for high versus low light collection efficiency crystals. The energy resolution for individual crystals varied between 14% and 23%. Partial detector arrays (e.g., 22/spl times/4) using 0.8/spl times/0.8/spl times/10 mm crystals with an etched surface finish have also been built and decoded. A miniature line source phantom consisting of five 1 mm diameter lines with 2 mm center-to-center spacing has been imaged using two partial MiCE2 detectors. All lines are distinguished and. the average peak to valley ratio between lines is 2.7.

Slat collimator design issues for dual head coincidence imaging systems

This paper investigates optimum slat collimator design parameters for dual-head coincidence imaging (DHCI) systems. The noise equivalent count (NEC) rate was examined with respect to the activity concentration under various system conditions. All results are derived from Monte Carlo simulations with a digital anthropomorphic (Zubal) phantom. The DHCI system was modeled after the Millennium VG gamma camera (GEMS, Waukesha, WI). The dead-time characteristics of the camera were experimentally determined. Our results suggests that substantial NEC gains can be achieved by varying the slat-to-slat separation, such that the peak of the NEC curve is located at clinically relevant levels (i.e., 0.07 /spl sim/ 0.10 /spl mu/Ci/cc). The NEC was also found to increase with the use of longer slats with appropriately selected slat-to-slat separation. Furthermore, the NEC performance also depends on the count-rate performance (i.e., dead-time losses) of the system. Therefore, as improvements are made to the count-rate capabilities of DHCI systems, the slat geometry should be modified. Further study is required to determine the effect that slat collimator design has on image quality and lesion detection for clinically realistic imaging situations.

Optimal Optical Conditions and Positioning Scheme for an Ultrahigh-Resolution Silicon Drift Detector-Based Gamma Camera

In this study, we optimized the optical conditions and associated positioning scheme for an ultrahighspatial-resolution, solid-state gamma detector. The detector module consisted of an array of seven hexagonal silicon drift detectors (SDDs) packed hexagonally and coupled to a single slab of crystal via a light guide glass. Because the optical behavior and requirements of the detector module and noise characteristics of the SDD sensor are different from those of conventional photomultiplier tube (PMT)-based detectors, the following parameters were studied to determine the optimum condition: scintillator selection, the effect of cooling on signal-to-noise ratio (SNR), the depth dependence of the scintillation light distribution, and optimum shaping time. To that end, a modified, Anger-style positioning algorithm with a denoise scheme was also developed to address the estimation bias (pincushion distortion) caused by the excessively confined light distribution and the leakage current induced by the SDD sensor. The results of this study proved that the positioning algorithm, together with the optimized optical configuration of the detector module, improves the positioning accuracy of the prototype detector. Our results confirmed the ability of the prototype to achieve a spatial resolution of about 0.7mm in full width at half maximum (FWHM) for 122 keV gamma rays under the equivalent noise count (ENC) of 100 (e- rms) per SDD channel. The results also confirmed NaI(Tl) to be a more desirable scintillator for our prototype with an energy resolution performance of about 8%.

Novel positioning method using Gaussian mixture model for a monolithic scintillator-based detector in positron emission tomography

We developed a high precision position decoding method for a positron emission tomography (PET) detector that consists of a thick slab scintillator coupled with a multichannel photomultiplier tube (PMT). The DETECT2000 simulation package was used to validate light response characteristics for a 48.8 mm×48.8 mm×10 mm slab of lutetium oxy- orthosilicate coupled to a 64 channel PMT. The data are then combined to produce light collection histograms. We employed a Gaussian mixture model (GMM) to parameterize the composite light response with multiple Gaussian mixtures. In the training step, light photons acquired by N PMT channels was used as an N-dimensional feature vector and were fed into a GMM training model to generate optimal parameters for M mixtures. In the positioning step, we decoded the spatial locations of incident photons by evaluating a sample feature vector with respect to the trained mixture parameters. The average spatial resolutions after positioning with four mixtures were 1.1 mm full width at half maximum (FWHM) at the corner and 1.0 mm FWHM at the center section. This indicates that the proposed algorithm achieved high performance in both spatial resolution and posi- tioning bias, especially at the corner section of the detector. C � 2011 Society

FPGA-based multichannel data acquisition system for prototype in-beam PET

In-beam positron emission tomography (PET) is a clinically proven imaging technique for the online investigation of position emitters induced by hadron irradiation. Because the PET involves the use of many photodetectors, the data acquisition (DAQ) system of the PET system requires many elements and a high processing speed to handle many input signals and complex data sets simultaneously. To create a fast and compact DAQ system for a prototype in-beam PET system, we used an FPGA-based multichannel analysis board and developed a firmware program that was geared to acquire the prototype in-beam PET data. As a result, a flood map of the detector block with 22Na 10μCi sources was acquired, and all pixels were well separated. 18F-FDG (4-mm diameter, 3 mCi), which was located in the center of a PET ring, was explored to acquire preliminary results from the prototype in-beam PET system. We produced an energy histogram with an energy resolution of 26% at 511 keV and a reconstructed point image. The measured maximum count rate on the host PC using the developed DAQ system was 120,000 cps. Although there were many improvements in terms of the count rate, the calculation of the pulse timing and correction methods can still be improved, and we were able to assess the feasibility of the prototype in-beam PET system from the achieved preliminary results.

A peak detection in noisy spectrum using principal component analysis

A spectrum of a radio isotope (RI) contains a single or multiple photo-peaks and radio-activities of all energy levels. These characteristics of each RI source are measured by radiation monitor (RM) systems. However, if the radiation count is extremely low and source to detector distance is too far, we cannot acquire good spectroscopic results for RI identification by RM devices while we still able to measure some counting statistics. Thus, precise peak detection in noisy spectrums is one of the most important tasks in the RM system. In this study, we developed an accurate peak detection method based on wavelet decomposition followed by principal component analysis. We used a discrete wavelet transform (DWT) for reduction of unnecessary high frequency noises in low counts spectrums. To reduce effect of a background radiation, we made a background template using a pre-measured background spectrum and calculated square errors for suppressing a background of low energy levels and maintaining true photo-peaks. Finally, we analyzed pre-processed data and detected photo-peaks using PCA. We measured Cesium 137(Cs-137) and Barium 133(Ba133) with 1 and 10 micro curies collected from the various distance. Each spectrum was collected for a second and total 60 sets were stored for each isotope. Results of our research show that the proposed algorithm achieves high sensitivity and specificity, proving that the algorithm is appropriate for RM systems.

Unmatched projector/backprojector pair for demultiplexing in multipinhole emission computed tomography

Statistically based iterative algorithms such as maximum likelihood-expectation maximization (ML-EM) are used for image reconstruction in single photon emission computed tomography (SPECT). Unmatched projector/backprojector pairs are sometimes used to accelerate the iteration process in the reconstruction algorithm. In this work, we propose and explore the use of an unmatched projector/backprojector pair for demultiplexing in multipinhole SPECT. Several simulations are conducted to evaluate the performance of the proposed method with uniform, hot-rod, and cold-rod phantoms. The proposed method incorporates an unmatched backprojector to utilize selective multiplexed projection data in reconstruction algorithms, while the projector is modeled as accurately as possible to represent realistic imaging geometry and the physical effects of multipinhole SPECT. The root mean square (rms) error and backprojection speed are evaluated to determine an unmatched backprojector. Our results demonstrate that the proposed method provides high-quality multipinhole SPECT images without multiplexing-related artifacts when a well-chosen unmatched backprojector is used.