Yuxuan Zhang

A new statistics-based online baseline restorer (SOBLR) for a high count-rate fully digital system

The goal of this work is to develop a novel, accurate, real-time digital baseline restorer using online statistical processing for a high count-rate digital system such as positron emission tomography (PET). In high count-rate nuclear instrumentation applications, analog signals are DC-coupled for better performance. However, the detectors, pre-amplifiers and other front-end electronics would cause a signal baseline drift in a DC-coupling system, which will degrade the performance of energy resolution and positioning accuracy. Event pileups normally exist in a high-count rate system and the baseline drift will create errors in the event pileup-correction. Hence, a baseline restorer (BLR) is required in a high count-rate system to remove the DC drift ahead of the pileup correction. Many methods have been reported for BLR from classic analog methods to digital filter solutions. However a single channel BLR with analog method can only work under 500 kcps count-rate, and normally an analog front-end application-specific integrated circuits (ASIC) is required for the application involved hundreds BLR such as a PET camera. We have developed a simple statistics-based online baseline restorer (SOBLR) for a high count-rate fully digital system. In this method, we acquire additional samples, excluding the real gamma pulses, from the existing free-running ADC in the digital system, and perform online statistical processing to generate a baseline value. This baseline value will be subtracted from the digitized waveform to retrieve its original pulse with zero-baseline drift. This method can self-track the baseline without a micro-controller involved. The circuit consists of two digital counter/timers, one comparator, one register and one subtraction unit. Simulation shows a single channel works at 30 Mcps count-rate with pileup condition. 336 baseline restorer circuits have been implemented into 12 field-programmable-gate-arrays (FPGA) for our new fully digital PET system.

Monte Carlo Simulation Study on the Time Resolution of a PMT-Quadrant-Sharing LSO Detector Block for Time-of-Flight PET

We developed a detailed Monte Carlo simulation method to study the time resolution of detectors for time-of-flight positron emission tomography (TOF PET). The process of gamma ray interaction in detectors, scintillation light emission and transport inside the detectors, the photoelectron generation and anode signal generation in the photomultiplier tube (PMT), and the electronics process of discriminator are simulated. We tested this simulation method using published experimental data, and found that it can generate reliable results. Using this method, we simulated the time resolution for a 13 times 13 detector block of 4 times 4 times 20 mm3 lutetium orthosilicate (LSO) crystals coupled to four 2-inch PMTs using PMT-quadrant-sharing (PQS) technology. We analyzed the effects of several factors, including the number of photoelectrons, light transport, transit time spread (TTS), and the depth of interaction (DOI). The simulation results indicated that system time resolution of 360 ps should be possible with currently available fast PMTs. This simulation method can also be used to simulate the time resolution of other detector design method.

A real time coincidence system for high count-rate TOF or non-TOF PET cameras using hybrid method combining AND-logic and Time-mark technology

A fully digital FPGA-based high count-rate coincidence system has been developed for TOF (Time of Flight) and non-TOF PET cameras. Using a hybrid of AND-logic and Time-mark technology produced both excellent timing resolution and high processing speed. In this hybrid architecture, every gamma event was synchronized by a 125 MHz system clock and generating a trigger associated with a time-mark given by an 8-bit high-resolution TDC (68.3 ps/bin). AND-logic was applied to the synchronized triggers for the real-time raw sorting of coincident events. An efficient FPGA based Time-mark fine-sort algorithm is used to select all the possible coincidence events within the preset coincidence time window. This FPGA-based coincidence system for a modular PET camera offers reprogrammable flexibility and expandability, so the coincidence system is easily employed, regardless of differences in the scale of the PET camera detector setup. A distributed processing method and pipeline technology were adopted in the design to obtain very high processing speed. In this design, both prompt and time-delayed accidental coincidences are simultaneously processed in real time. The real-time digital coincidence system supports coincidence in 2 to 12 detector module setups, capable of processing 72 million single events per second with no digital data loss and captures multiple-event coincidence for better imaging performance evaluation. The coincidence time window-size and time-offset of each coincidence event pair can be programmed independently in 68.3 ps increments (TDC LSB) during the data acquisition in different applications to optimize the signal-to-noise ratio. The complex coincidence system is integrated in one circuit board with 1.5 Gbps fiber optic interface. We demonstrated the system performance using the actual circuit and Monte Carlo simulations.

New 9

Our objectives were to develop two 9times9 and 10times10 array high resolution position sensitive block detectors using low cost photomultiplier-quadrant-sharing (PQS) technology. These two blocks can decode 4.3 mm or 3.9 mm BGO crystals with 39 mm size regular round PMT. By using ESR film as reflector between crystals, we achieved 96% packing fractions for these two blocks. Compared with conventional block design using 19 mm or 26 mm PMT, Phi39 mm size PMT can reduce the number of PMT in a camera by 55%-75%. Use of BGO crystal can reduce crystal cost by more than 70% compared to LSO or GSO crystals that are used in some of the newest PET cameras. The BGO crystal sizes-4.3 mm or 3.9 mm-in these two PQS blocks were similar to or smaller than BGO, LSO, or GSO crystals (4-6.3 mm) in commercial human PET cameras. The spatial resolution of the PQS blocks was expected to be similar or better. List mode data of these two blocks were acquired with Na- 22 source. Crystal-decoding map of blocks and individual crystal spectra were derived. All 9times9 and 10times10 crystals were clearly decoded on the block decoding map. These two PQS detectors decoded 81 or 100 BGO crystals per PMT. The average peak to valley ratio of decoding map was 3.4 and 1.8 respectively. The light collection efficiency for all crystals in the blocks were 71%-100% and 66%-100% respectively. The average energy resolutions of all crystals were 16.8% and 17.2% respectively. Our data suggest that even at lower cost the new PQS PET camera may outperform recent conventional PET cameras.

\,\times\,

Time-of-flight (TOF) data can improve the imaging quality of positron emission tomography (PET) cameras by estimating the location of the positron annihilation. In this study, we measured the TOF resolution of lutetium yttrium orthosilicate (LYSO) blocks based on the photomultiplier-quadrant-sharing (PQS) design, which has no external light guide and therefore has higher light collection efficiency than the conventional block design with a light guide. We investigated three digital timing methods to achieve better timing resolution: constant fraction discrimination, leading edge discrimination, and linear fitting. A time-energy correction, estimated from a least mean square fit, was applied to minimize time walk caused by the variation of signal shapes. The average block-to-block coincidence timing resolution for two 13 ×13 blocks, each made of 4 × 4 × 20 mm3 LYSO crystals, with fast R9779 PMTs from Hamamatsu (51-mm diameter) was estimated to be 422 ps (full width at half maximum). Another 13 × 13 block of smaller 1.4 × 1.4 × 10 mm3 LYSO crystals coupled to regular 19-mm-diameter XP1912 PMTs from Photonis was also tested and an average block-to-block timing resolution of 527 ps was achieved. This study shows that with PQS detector blocks, good time resolution suitable for TOF PET systems can be obtained, as well as high spatial resolution and lower PMT costs.

9 and 10

A high resolution transformable PET camera for human research has been developed. It can be transformed from whole-body mode (83 cm detector ring diameter, 13 cm axial FOV) to brain/breast mode (54 cm diameter, 21 cm axial FOV). This flexibility allows more optimal imaging for more dedicated human research studies, compared to commercial systems that are designed primarily for whole-body cancer imaging. The camera has 12 rectangular detector modules with gaps between modules which are small and remain constant for every configuration. Each module has a large detection area of 13 times21 cm2 and 3,168 BGO detector crystals of 2.68 times 2.68 times 18 mm3 with built-in front-end electronics. The system has 38,016 BGO detectors and 924 PMT in total. The low-cost high-resolution PMT-quadrant-sharing (PQS) detector design was used for the detection system. A fast LED cross-reference calibration method is used to maintain the PMT gain balance. To improve the spatial sampling, the gantry rotates 30 with a fine step of 1along with the high-speed electronics and detector modules. The whole system (detectors, electronics, mechanics, and software) was designed and fabricated inhouse. The performance characteristics are measured following the NEMA NU 2-2001 standard. 3D acquisitions were collected without interplane septa in the system. The sensitivity is 4.2% for whole-body mode and 9.2% for brain/breast mode with a Na-22 point source placed at FOV center. For brain/breast mode, the transaxial image resolutions at 0, 10 cm are 2.7 and 4.0 mm; while for whole-body mode, the resolutions are 3.3 and 3.9 mm. The slice thickness resolution is 2.6 to 3.3 mm in brain mode. DOI reduction technique is applied into the sinogram with the model- based Point Spread Function (PSF) obtained by Monte Carlo simulations to achieve high-definition (HD), high uniform PET images. The improved image resolutions are 2.9, 3.3 mm in whole-body mode at the location of 0, 10 cm offset. A coincidence window of 13 ns collects 93% of the trues. The scatter fraction is 48% in brain mode. The maximum noise- equivalent count rate (NECR) is 34 Kcps at 4.6 KBq/cc for the brain mode and 20 Kcps at 5.6 KBq/cc for whole-body mode with the 70-cm NEMA NECR phantom. The convertible lead shield is being redesigned for higher NECR in whole-body mode.

\,\times\,

We produced 40,000 high-resolution position-sensitive BGO detectors with a new method (slab-sandwich-slice or SSS). The detectors will be used in a high-resolution human PET system. The detectors were 2.68/spl times/2.68 mm (18 mm deep) encapsulated in about 800 detector blocks. There were three types of blocks/arrays made, 7/spl times/7, 7/spl times/8, 8/spl times/8. We studied the optical and physical characteristics of the production to evaluate the SSS production method. We also analyzed the data relative to production experience and time. We found that for the 7/spl times/7 symmetric blocks: the average block-composite energy resolution was 35%; the average individual-crystal energy resolution was 23%. Mean intra-block light-collection variation of individual crystals was 23% FWHM. The average number of clearly-decoded crystals was 47 out of a possible 49 for the 7/spl times/7 block in the early production, which was improved to 49 in late production. For the 7/spl times/8 asymmetric blocks, the average blocks-composite energy resolution was 32%. Mean intra-block light-collection variation was 23% FHWM. The average number of clearly decoded crystals was 51 for the early production improving to 55 for the late production (out of a possible 56). For the 8/spl times/8 asymmetric blocks, the average blocks-composite energy resolution was 29%. Mean intra-block light-collection variation was 18% FHWM. The average number of clearly decoded crystals was 59 in early production improving to 62 for the late production (out of 64). We have also studied other production quality issues and found production variability and imperfection in the areas of size, shape and color. We observed a (a) darkness effect that for a block with poor crystal-position decoding map there was generally darker regions within the blocks and several size and shape errors were also found, as (b) X-Y thickness-difference effect, which is the difference between the axial and transaxial block dimensions, (c) pyramid-effect with the PMT-end of the block (top) and the patient-end of the block (bottom) had different sizes, (d) besides the pyramid-effect, the 4 sidewalls might not orthogonal due to a systematic lateral shift between slices, or laminar-shift dislocation effect (e) the slice-serration effect which is caused by the variation in slice-widths between different sandwich types, (f) paint-edge effect, where there was paint material build-up at the edge of the painting mask.

10 BGO Block Detector for Human PET Using PMT Quadrant Sharing Design

Using standard circular photomultipliers in quadrant sharing configuration (PQS), four new high resolution L(Y)SO detector blocks for human PET applications have been developed. The detector blocks were assembled in very large arrays of 14times14, 15times15, and, 16times16 for two of them. The crystal pitch sizes were 2.74times2.74times20 mm3, 3.25times3.25times20 mm3, 2.42times2.42times20 mm3, 2.78times2.78times20 mm3 and the surface finishing was 4-mum. Mirror films patterns were placed between crystals as reflectors. The blocks were assembled using optical grease and wrapped by Teflon tape. To acquire list mode data of Ga-68 source (511 KeV) with at high voltage bias of 1100 V, these blocks were coupled to regular round PMTs of 39/51 mm and, crystal decoding maps, individual crystal energy resolutions and relative light-collection efficiencies were extracted from this data. To investigate the potential imaging resolution of the human PET scanners with these blocks, GATE (Geant4 application for tomographic emission) simulation package was used. The packing fractions of these blocks were found to be 97%, 97% and 95%-97%. From decoding maps, all 196, 225 and 256 crystals were clearly identified. The average energy resolutions were 14.3%, 15.5% and 13.1%/14.2%. The relative collection efficiencies were 0.81~1.00, 0.83~1.00 and, 0.77~1.00/0.78~1.00. From this study we can conclude that the PQS design has the potentiality to achieve high spatial resolution and excellent energy resolutions on human PET scanners with substantially lower production cost.