Niraj K. Doshi

Monte Carlo simulation study of LSO crystals

In order to study light transport in LSO crystals for PET, we simulated single crystals and block detectors using Detect2000. We devised basic experiments to evaluate fundamental physics parameters of the simulation and benchmarked the simulation with a set of experiments on single crystals. The block detectors that were studied were composed of Lutetium Oxyortho-Silicate crystals, a light guide and Photo Multiplier Tubes. The results of the simulation were compared to an experimental detector blocks position profile. By refining our simulation to model actual detectors, a tool was created that can be used to optimize the performance of detector blocks.

Comparison of typical scintillators for PET

There has been a plethora of literature describing the various properties of scintillators that are commonly used in PET detectors. In this literature there usually is a comparison made with respect to light output and energy resolution. Unfortunately, any of these comparisons are misleading because the treatment of the different scintillators is not optimized to produce the best results that may, be possible through proper surface treatment of the scintillator. In this research, there is a comparison made between the following scintillators: BGO, LSO, GSO, LYSO, and YSO. Again, the figures-of-merit for comparison were light output and energy resolution at 511 keV.

Quarter-Trio Mapping Electronics Readout Scheme for APD Block Detector in PET

Quarter-trio mapping (QTM) is a new detector readout scheme for avalanche-photodiode (APD) block detectors in positron emission tomography (PET). This design simplifies the detector readout electronics, and more efficiently utilizes APDs in detector blocks hence reducing cost. With the QTM design, eight APDs are positioned in each detector block. The APDs are positioned in such a way that 12 times 12 crystal elements each of 2.5 mm times 2.5 mm times 20 mm can be identified. Moreover, we developed an efficient electronics readout QTM scheme in which only four output channels instead of eight are sent to main electronics to readout eight APD signals. Compared with nine APDs in a 3 times 3 array setup, initial results from eight APDs in QTM readout shows that the 12 times 12 crystal block can be well identified. The timing resolution (crystal block of 12 times 12 elements vs. single plastic scintillator) is about 2.95 ns vs. 2.84 ns from a 3 times 3 crystal array. More experiments are on going to further improve the energy and timing performance.

A pulse shape restore method for event localization in PET scintillation detection

Localizing gamma-ray events accurately and performing pileup rejection/correction functions are desirable in positron-emission-tomography (PET) front-end electronics development. Two techniques, the traditional analog integration with charge-sensitive amplifiers and the recent digital integration by using free-running analog-to-digital converters (ADCs), are the typical methods to obtain the event energy and position information. Pileup issues have been extensively investigated in both these techniques. In this new study, a pulse-shape-restore (PSR) method for event localization is presented. From each PET scintillation detector, a photo-sensor current output signal is amplified then conditioned by a filter. Subsequently the signal is digitized with a fast sampling free-running ADC. The digitized signal is finally processed in a Field Programmable Gate Array (FPGA) by using a numerical line fitting method to restore the signal to its theoretic shape. The event energy is directly obtained from the restored pulse shape rather than from the integration calculation. With the PSR method, we may enhance the event localization accuracy and improve the signal energy resolution. Moreover, the PSR method will be implemented as a pileup rejection /correction algorithm to improve the detector count-rate ability and reduce the gamma ray mispositioning in high count-rate conditions

(A, B, C, E) and (T, L, E) Multiplexing Readout Concept for PET Block Detectors

New A, B, C, E (energy) and T (top), L (left), E (energy) multiplexing readout concepts for block detectors in positron emission tomography (PET) have been developed. With the ABCE (TLE) design, eight (four) detector blocks with 32 (16) channels can be multiplexed into 4 (3) outputs. The new multiplexing scheme is capable of substantially reducing the signal processing channels in the data processing main electronics. In addition, the new scheme could improve system timing performance by eliminating cable time-skew and facilitate the filter design by downgrading the circuit accuracy requirements such as the group-delay error and filter channel skews. As a drawback, the processing channel reduction will add to detector pile-up and dead-time. These effects need to be carefully investigated for any implementations.

On the efficiency of NaI/LSO and YSO/LSO phoswich block detectors

In order to obtain a high and uniform spatial resolution in PET, the depth-of-interaction information (DOI) has to be established. This information can be obtained by using a phoswich detector, such as GSO and LSO or other scintillator combinations. The authors are looking into efficiency questions concerning phoswich detectors. As a part of this investigation, NaI(Tl)/LSO and YSO/LSO phoswich block pairs have been investigated in both singles mode and in coincidence counting mode in order to measure the counting efficiency for 511 keV annihilation photons. The reason for having a soft scintillator in front is to be able to switch modality, combing both PET and SPECT in the same detector. When operated in PET mode, however, both scintillator layers are used. The results were compared to LSO only data, the LSO detector being the same as was used in the phoswich block. The two phoswich blocks were investigated both experimentally and theoretically. A satisfactory agreement was found between the experiments and theory, with the latter based on simulations with GEANT. The phoswich combination is, as expected, more effective than the LSO only detector due to intra-detector events. For a soft scintillator like NaI or YSO, however, the phoswich gain is small for 511 keV photons with an energy threshold above 350 keV in both scintillators.