P01 — Sunday, October 16, 2011, 16:00 — 16:30, NIA Exhibition Area Physics & Instrumentation & Data Analysis: Instrumentation P001 Comparison of three techniques for calculating the 20% count rate loss in a gamma probe for sentinel node biopsy surgery K. L. Kirkbride, A. G. Irwin, T. Hague, A. J. Craig, A. Britten; St George's Hospital, London, UNITED KINGDOM. AimTo compare two standard methods (“decaying source” and “two-source” methods) for measuring count rate performance in gamma probes with a new method (“distance method”), and to evaluate the tests for their suitability as part of routine QA procedures.IntroductionCharacterisation of the count rate capability due to dead time is a component of gamma probe QA. Dead time reduces the recorded counts and QA measurements must be taken on the linear response range. NEMA recommends the decay method for gamma probes (1), whilst the two source method is a standard technique for nuclear instrumentation (2). No method is ideal. The decaying source method requires measurements over a long time interval, whilst the two-source method requires the count rate loss value to be roughly known, and it is an indirect measurement. The proposed distance method has the potential to be rapid and to be workable direct measurement with little a priori knowledge required.MethodThe 20% count rate loss count rate (R 20 ) of a Europrobe gamma probe was estimated using the standard decaying source and two-source methods. The new distance method was applied using a “hot” source and a “cool” source and changing the count rate by varying the source-probe distance. The two responses were compared using the cold source count rate as a reference to obtain the R 20 value.ResultsThe 20% count rate drop R 20 was calculated to be at 3,937 cps (decay method), 3,500 cps (two-source method), 3,991 (distance method). The decay method duration was about 12 hours, whilst the two-source and distance methods took about an hour to prepare sources and record measurements, and both need a jig and/or water bath.ConclusionsGood agreement was observed between the decay method and the proposed distance method. This may be a practical, shorter-time alternative to the established decay method and further comparisons are being undertaken into reproducibility between probes and operator variability.References1. NEMA standards publication NU3-2004 “Performance measurements and quality control guidelines for non- imaging intraoperative gamma probes,” 20042. Adams et al., “Deadtime measurements in scintillation cameras under scatter conditions simulating quantitative nuclear cardiography,” J Nucl Med 19: 538-544, 1978 P002 Design and Construction of Shielding Sleeves to Test Radionuclide Calibrator Linearity for Clinical Use D. N. B. Yatigammana, J. L. Merrett, A. H. McLean, M. A. Masoomi; Queen Alexandra Hospital, Portsmouth, UNITED KINGDOM. Aim To investigate and design multi-element shielding sleeves to use as an alternative to the decaying source method of performing linearity measurements for radionuclide calibrators. This will allow linearity QC to be performed in a matter of minutes by replicating the natural decay of the source over a few days.Materials and Methods A design was created by using a line source model to calculate the range of multi-element thicknesses needed to shield a Tc-99m source to produce a range of measurements across the entire range of clinically assayed activities. The sleeve design was chosen to ensure consistent geometry between measurements and various designs of sleeves were analysed to assess the range of measurements that could be performed and the practicality (weight, cost, etc) in performing the measurements. We investigated the attenuation properties of various materials when considering the final design.As part of the initial analysis, existing lead pots were used to determine the constancy in assays of a Tc-99m source through lead.Results The range of Tc-99m activities used in the design of the sleeves was 10 - 6000 MBq to match the clinical range assayed. A sleeve set was designed such that combinations of sleeves could be placed within each other to produce 12 measurements with 4 sleeves, minimising the overall time of measurements and the manufacturing costs. The initial measurements through existing lead pots showed that there was consistency in the method of measurement with a standard deviation of 2.6% in the calculated equivalent decay time.Conclusion Using the initial findings, a final design using alternative materials and manufacturing methods will be considered. A test design will be produced to measure the difference between expected and measured attenuation, considering scatter contribution, to produce a more accurate model with which to finalise the design of the sleeves. P02 — Sunday, October 16, 2011, 16:00 — 16:30, , NIA Exhibition Area Physics & Instrumentation & Data Analysis: Image Reconstruction P003 Improvement of image quality using cross projection angle data in emission computed tomography K. Matsumoto 1 , K. Shimizu 2 , K. Murase 3 , Y. Wada 4 , Y. Watanabe 4 , M. Senda 2 , T. Mukai 1 ; 1 Kyoto College of Medical Science, Kyoto, JAPAN, 2 Institute of Biomedical Research and Innovation, Kobe, JAPAN, 3 Graduate School of Medicine Osaka University, Osaka, JAPAN, 4 RIKEN Center for Molecular Imaging Science, Kobe, JAPAN. Aim: General data acquisition for single photon emission computed tomography (SPECT) is performed in 90 or 60 direction using a gamma camera. On the other hand, positron emission tomography (PET) is measured from 128 to 256 directions, but cannot be changed as one chooses. The purpose of this study was to investigate a cross projection angle (CPA) method of emission computed tomography (ECT) by applying the cross plane in PET.Method: The CPA method was evaluated to a digital phantom (Shepp-Logan phantom) with attenuation and statistical noise and a Hoffman 3-D brain phantom. A PET scanner and a gamma camera, CTISiemens ECAT EXACT HR+ and GE Healthcare Millennium MG, were used to scan images of a Hoffman 3-D brain phantom. The projection data of CPA method was corrected for normalization. Attenuation and scatter corrections of SPECT image were performed using Chang's method and a modified triple energy window, respectively. Each correction (i.e. attenuation, scatter, deadtime, etc.) of PET image were used recommended for the scanner model. All ECT images were reconstructed by the filtered back-projection method with a ramp filter cut off at the Nyquist frequency. Quality of the reconstructed images was evaluated in terms of Normalized Mean Square Error (NMSE) values and a two dimensional power spectrum analysis.Results: NMSE values of PET and SPECT images using CPA method were very nearly zero. As for a two dimensional power spectrum analysis, PET and SPECT images quality using CPA method were approximately 1.8 times better than conventional method. Also, the spatial resolution of CPA method was not change compared with the conventional method.Conclusion: Our preliminary data suggest that effective improvement of image quality can be expected with cross projection angle. Also, the CPA method was considered clinical useful and a widely available simple method. P004 The role of reconstruction algorithm and SPECT acquisition orbits in somatostatin receptors’ detectability F. Voltini, F. Zito, G. Marotta, R. Lambertini, C. Canzi, M. Migliavacca, P. Gerundini; Nuclear Medicine Department, IRCCS Fondazione Ca’ Granda Ospedale Maggiore, MILAN, ITALY. AimThe aim of this work was to evaluate the effects on the performance in somatostatin receptors’ detectability due to both reconstruction algorithms and different acquisition orbits.Material and MethodsThe study was performed by the ɶ-camera SYMBIA-S(Siemens), simulating SPECT patients examinations using an anthropomorphic Alderson phantom with seven inserts of different volumes(0.4÷4 ml), to reproduce tumoral foci in thorax, liver, abdomen. Mediastinum, liver, abdomen and lesions were filled with 111 In water solution with different concentrations to simulate lesion-to-background ratios found in clinical SPECT studies. Acquisitions were performed by body-contour(BC) and circular(CO) orbits. Data were reconstructed by FBP and iterative algorithms implemented on e.soft- workstation:Osem-2D and Flash-3D TM . Flash-3D TM includes 3D-beam-collimation model. Iterative reconstructions were performed by 20 sub-sets and 5,10,15,20,30 iterations;the choice of sub-sets was previously optimized. Eleven sets of images were reconstructed for both OC and BC studies.Lesion detectability was assessed by contrast(C%) and signal-to-noise-ratio(SNR) therefore all the sets of images were analyzed: 1) to calculate C% and SNR for each lesion in the phantom acquired with circular orbit (C%L CO ; SNR-L CO ) and body-contour(C%L BC ;SNR-L BC ); 2) to evaluate the paired t-test between correspondent values of C%L CO ,C%L BC and SNR-L BC , SNR- L CO .ResultsC%L CO (C%L BC ) evaluated on the images reconstructed with Flash-3D TM , were significantly higher (p<0.05) than correspondent values C%L CO (C%L BC ) calculated with OSEM-2D and FBP, similarly for the C%L CO (C%L BC ) of OSEM-2D compared to correspondent values of FBP. For each reconstruction algorithm differences between C%L CO and C%L BC were not significant.The mean values of C%L CO were for Flash-3D TM , OSEM-2D and FBP respectively: 76,72,57 while for C%L BC were respectively: 75,73,59.SNR-L BC (SNR-L CO ) values for Flash-3D TM were, both OC and BC, significantly lower (p<0.05) than correspondent values SNR- L BC (SNR-L CO ) for OSEM-2D and FBP. The differences between SNR-L BC and SNR-L CO values were significant only for OSEM-2D.ConclusionsPerformance in lesion detectability are influenced mainly by the choice of reconstruction algorithm rather than by the kind of orbits used for SPECT. When assessed for Flash-3D TM contrast was significantly higher, for both OC and BC, than OSEM-2D and FBP while the POSTER PRESENTATIONS Eur J Nucl Med Mol Imaging (2011) 38 (Suppl 2):S260–S441 DOI 10.1007/s00259-011-1 - 911 0