IEEE 2017 NSS/MIC/RTSD ControlCenter

Online Program Overview Session: J-02

To search for a specific ID please enter the hash sign followed by the ID number (e.g. #123).


Session chair: Micah Folsom; Eric Frey Johns Hopkins University
Shortcut: J-02
Date: Wednesday, October 25, 2017, 10:20
Room: Centennial III
Session type: Joint Session


10:20 am J-02-1

High Speed Dose Imaging Detector for hadron-therapy based on Well-type Glass-GEM (#2260)

T. Fujiwara1, Y. Koba2, Y. Mitsuya3, H. Toyokawa1, H. Takahashi3

1 National Institute of Advanced Industrial Science & Technology (AIST), Tsukuba, Ibaraki, Japan
2 National Institute of Radiological Sciences (NIRS), Chiba, Chiba, Japan
3 The University of Tokyo, Bunkyo, Tokyo, Japan


Hadron therapy is known as one of the most efficient radiation therapies for cancers. For daily quality assurance (QA) measurements in hadron (mostly proton and carbon) radiotherapy, a dosimetry system that has a two-dimensional effective area, high spatial resolution, and linear response-to-dose is required. In addition, active spreading, which is known as spot-scanning-treatment is another today’s important technique for hadron therapy.

In this study, we demonstrate real-time dose-imaging performance of our high digital dose-imager with Proton/Carbon beam for hadron therapy. The imager based on newly developed gaseous detector, Well-type Glass-GEM. The imaging system is successfully operated in hadron therapy facility (HIMAC) with practical intensity beam for radiotherapy. It shows high spatial resolution (< 1 mm) and very low Linear-energy-transfer (LET) dependence. In addition, high speed image forming is demonstrated with taking spot-scanning beam movies at frame rate 10 fps.

Keywords: GEM; Hadron therapy; dose-imaging; Glass GEM; quality assurance
10:38 am J-02-2

A novel method for temporally separating Cherenkov radiation in a scintillator probe in x-ray therapy (#1846)

J. Archer1, L. Madden1, E. Li1, M. Carolan3, M. Petasecca1, 2, P. Metcalfe1, 2, A. Rosenfeld1, 2

1 University of Wollongong, Centre for Medical Radiation Physics, Wollongong, New South Wales, Australia
2 University of Wollongong, Illawarra Health and Medical Research Institute, Wollongong, New South Wales, Australia
3 Wollongong Hospital, Illawarra Cancer Care Centre, Wollongong, New South Wales, Australia


Cherenkov radiation is the primary source of light contamination in an optical system exposed to high-energy radiation. Therefore the removal of the contribution of Cherenkov radiation to a light signal is vital, especially in the field of dosimetry where accurate measurements of radiation dose is critical for the quality assurance of radiation sources and treatment plan validation. We present and demonstrate a novel method for Cherenkov subtraction in a plastic scintillator fibre optic dosimeter applicable to regular, pulsed radiation sources. BC-444 plastic scintillator is used because of its long rise time of 19.5 ns, allowing an algorithm to analyse the total light output waveforms from the dosimeter to discriminate between the Cherenkov light contribution to the signal and the scintillation contribution. This discrimination relies on the Cherenkov radiation reaching its maximum output much before the scintillator does. A beam profile and percent depth dose of a 6 MV x-ray beam was measured using the dosimeter, and analysed using the presented algorithm. The algorithm correctly identified the time when the Cherenkov signal saturates, however the results over-respond when compared to ionisation chamber data, due to the beam intensity, and hence Cherenkov light, increasing consistently for 1.5 μm after the beginning of the beam pulse. This results in the Cherenkov contribution being underestimated using the presented method. Further characterisation of the Cherenkov light signal at different dose levels has the potential to allow a simple correction factor to me implemented to correct the over-response.

Keywords: scintillators, cherenkov radiation, x-ray, dosimetry
10:56 am J-02-3

Evaluation of a FPGA-based Real-Time Coincidence Module for High Count Rate PET Scanners (#1878)

J. - W. Son1, 2, J. Y. Won1, 2, J. S. Lee1, 2

1 Seoul National University, Department of Nuclear Medicine, Seoul, Republic of Korea
2 Seoul National University, Department of Biomedical Sciences, Seoul, Republic of Korea


In this study, we present a timestamp-based real-time coincidence system that is implemented in a field-programmable gate array (FPGA). The coincidence module was designed as a pipeline and consisted of shift registers and a combinational logic. Events are classified as singles, coincidences, or multiples by a combinational logic within one clock (i.e., 8 ns) without losing data. By using shift registers in the coincidence system, unpaired singles can be effectively rejected because every event waits for a possible coincidence event for only about 50–200 ns (i.e., clock periods × depth of the module).

To evaluate the performance of the coincidence system, we downloaded data generated from GATE Monte Carlo simulation toolkit to a FPGA evaluation board and compared the number of coincidence pairs to the real coincidence results. A small-ring PET scanner for specific organ-dedicated imaging was simulated. The average error rate of the acquired number of prompt coincidences was only 0.002% in the nine different simulation studies (activity concentration within a uniform phantom from 0.73 kBq/ml to 29.1 kBq/ml), which demonstrated the validity of the developed coincidence system.

To investigate the count rate improvement of a real PET scanner, we measured noise equivalent count rate (NECR) of a prototype time-of-flight PET scanner with and without applying the coincidence system. The peak NECR and true count rate were improved by 130% and 137%, respectively by using the coincidence system. The source activity concentration of the phantom at NECR peaks were 177% and 231% higher.

The results demonstrated the validity and performance of the developed coincidence system. The module can handle a larger number of detector channels without changing its architecture, and therefore we can use it for different scanner geometries. We will apply the coincidence system to our brain-dedicated PET scanner under development, and present its count rate performance.

Keywords: data acquisition, positron emission tomography, coincidence system
11:14 am J-02-4

Signal walk correction for high-precision TOF measurement (#4133)

Q. Peng1, J. Qi2, W. W. Moses1, Z. Zhao3, Q. Huang3, J. Xu4

1 Lawrence Berkeley National Laboratory, Berkeley, California, United States of America
2 University of California, Davis, Davis, California, United States of America
3 Shanghai Jiaotong University, Shanghai, China
4 Huazhong University of Science and Technology, Wuhan, Hubei, China


Leading edge discriminator (LED) is triggering method commonly used in the readout electronics in the TOF PET system. One drawback of LED is that the timing measurements are affected by the signal walk. The purpose of this study is to investigate the effects of the signal walk in the timing measurement and the method to compensate it.

The experimental results from a pair of LaBr3 detectors show that: (1) the signal walk causes a logarithmic relation between energy and measured time. A logarithm operation need to be applied to the energy and before the linear correction can be applied. We named this correction Qi-correction. (2) The signal walk dramatically affect the FWHM CTR of the detectors when the lower bound of the energy window is high. The measured FWHM CTR increased by a factor of 6 to 561 ps from 91.5ps, when the lower cut of the energy window decreased from 450 keV to 100keV. (3) The Qi-correction can dramatically improve the CTR. The FWHM CRT with energy window [100keV 560keV] improved by a factor of 3.9 from 561ps to 145ps after the Qi- correction is applied.

We conclude that it is essantial to apply the Qi-correction when constructing high-performance TOF PET systems with the dual-end readout detectors, or monolithic scintillators, or scintillator with low photofraction rate (for example, LaBr3).

Keywords: Signal walk correction, Coincidence timing resolution, Analog SiPM, Time of Flight
11:32 am J-02-5

Evaluation of a TOF-PET detector design that achieves ≤100 ps coincidence time resolution (#3885)

J. W. Cates1, C. S. Levin1

1 Stanford University, Radiology, Stanford, California, United States of America


Positron emission tomography (PET) detectors for clinical systems employ scintillation crystal element that are typically ~20 mm length and 3-5 mm in width optically coupled on their narrow end to a photosensor. The aspect ratio of this traditional crystal element configuration yields low light collection efficiency of scintillation photons and significant, interaction depth-dependent scintillation light transit time jitter from the 511 keV photon interaction position to the exit interface of the crystal. Alternatively, coupling the photosensor on the long side of the crystal elements results in near-complete light collection efficiency and low scintillation photon transit time jitter. In this new scintillation light readout configuration, the achievable coincidence time resolution (CTR) with 3x3x20 mm3 LGSO:Ce(0.025 mol%) crystals coupled to a row of SensL-J SiPMs, multiplexed to a single timing channel, is 102±2 ps FWHM using simple leading edge time pickoff. This is in contrast to a CTR of 137±2 ps FWHM that is achieved when the same crystals are coupled to a single 3x3 mm2 SiPM in the standard "end coupled" configuration. In this work, we present a theoretical and experimental investigation into the achievable timing performance using this side readout configuration. We further study the statistical limit on CTR using side readout via the Cramér-Rao lower bound (CRLB), with consideration given to ongoing work to further improve photosensor technologies and exploit fast phenomena to ultimately achieve 10 ps FWHM CTR. Altogether, the side readout configuration offers an immediate solution for 100 ps clinical PET detectors and mitigates factors affecting ongoing efforts to achieve 10 ps FWHM TOF-PET.

Keywords: TOF-PET
11:50 am J-02-6 Download

INSERT: A novel clinical scanner for simultaneous SPECT/MRI brain studies (#2031)

B. F. Hutton1, K. Erlandsson1, D. Salvado1, M. Occhipinti2, 11, Z. Papp3, J. Willems4, C. Piemonte5, M. Carminati2, 11, T. Bukki3, A. Kühne6, Z. Nyitrai3, T. Niendorf6, P. van Mullekom4, H. Waiczies6, K. Nagy3, I. de Francesco7, D. Mathe8, L. Ottobrini9, S. C. Short10, C. Fiorini2, 11

1 University College London, Institute of Nuclear Medicine, London, United Kingdom of Great Britain and Northern Ireland
2 Politecnico di Milano, Dipartimento di Elettronica Informazione e Bioingegneria, Milan, Italy
3 Mediso Ltd., Budapest, Hungary
4 Nuclear Fields, Vortum-Mullem, Netherlands
5 Fondazione Bruno Kessler (FBK), Trento, Italy
6 MRI.TOOLS GmbH, Berlin, Germany
7 University College Hospital, Department of Oncology, London, United Kingdom of Great Britain and Northern Ireland
8 CROmed Ltd., Budapest, Hungary
9 University of Milan, Dept. of Medical-Surgical Pathophysiology and Transplants, Milan, Italy
10 University of Leeds, Faculty of Medicine and Health, Leeds, United Kingdom of Great Britain and Northern Ireland
11 INFN, Sezione di Milano, Milano, Italy


A clinical SPECT insert for a commercial MRI scanner has been developed within the INSERT project, allowing simultaneous SPECT/MRI studies of the human brain. The INSERT system consists of a partial ring of 20 scintillation detectors with CsI(Tl)-crystals coupled to 5x10 cm2 SiPM-readout with an intrinsic resolution of 1 mm FWHM. The scanner is stationary and is equipped with multi-mini-slit-slat tungsten collimators, and a dedicated transmit/receive RF-coil. This work presents the last stages of the INSERT assembly and preliminary experimental performance of the system using a single prototype camera and a rotating platform. List-mode data were acquired for various phantoms: Four capillary tubes placed at different axial distances from the FoV centre, seven spheres with different sizes and activity concentrations, a uniform cylinder with a 184 mm diameter, the Hoffman brain phantom and the Alderson striatal phantom. The first 4 were filled with 99mTc, and the last one with 123I. The sensitivity was measured with a small cylindrical source at 61 positions in the FoV. Positions of the detected events were determined by a ML-EM algorithm based on estimated light-spread functions. Images were reconstructed with a ML-EM or MAP-EM algorithm using a projector/back-projector based on angular blurring, after model-based geometrical calibration of the acquired data. Reconstructed images of the brain phantoms show expected activity distribution. Quantitative analysis of activity concentration of the spheres and striatal compartments show correct relative concentrations. The reconstructed resolution is <10 mm, and improves from the centre towards the edge of the FOV. The cylinder phantom showed reasonable uniformity. The system sensitivity was estimated to 1.5 10-4 s-1/MBq. In conclusion, preliminary experimental evaluation with a single camera shows good qualitative and quantitative results. The INSERT is the first clinical SPECT prototype for simultaneous SPECT/MRI.