Physica Medica: European Journal of Medical Physics - Volume 32 - Supplement 1 - February 2016

107 / 146

C.346
PRELIMINARY ASSESSMENT OF RADIONUCLIDIC PURITY OF CYCLOTRON
PRODUCED 99MTC
G. Cicoria A. Corazza F. Zagn S. Vich D. Pancaldi G. Pupillo
A. Boschi P. Martini L. Uccell M. Pasqual A. Duatti G. Di Domenico
M. Loriggiola M. Bello J. Esposito M. Marengo a Policlinico S.Orsola-
Malpighi, Bologna, Italy; b INFN, Legnaro (PD), Italy; c University of Ferrara,
Ferrara, Italy; d University of Padova, Padova, Italy
Introduction: Cyclotron production is a promising, alternative route to
supply 99mTc with high yields, potentially attainable with cyclotrons usually
dedicated to PET isotopes production. As described in the recent draft of
Pharmacopoeia Monography, the only significant difference between
generator- and cyclotron-produced 99mTc is the radionuclidic impurity
profile. The aim of this work was the preliminary assessment of radionuclidic
purity of 99mTc produced with a 16.5 MeV GE PETtrace cyclotron at the
University Hospital of Bologna.
Materials and Methods: The irradiation of a target, consisting of 5 foils
of enriched 100Mo (92Mo 0.08%, 94Mo 0.05%, 95Mo 0.10%, 96Mo 0.11%,
97Mo 0.07%, 98Mo 0.54%, 100Mo 99.05%), produced by powder lamina-
tion at the LNL-INFN (Legnaro), was conducted with a proton beam current
of 20 microA with an entrance energy of 15.7 MeV for irradiation time of
5400 s. The chemical separation and extraction were performed by a fully-
automated module, developed in collaboration with the University of Ferrara.
Radioisotopic content in final formulation and in the residuals after ex-
traction procedure was quantified using gamma-ray spectrometry. Samples
were analyzed at the end of synthesis (EOS), and after 12 h, 24 h and 48 h.
Results: The evaluated saturation yield was 1.1 GBq/microA. Technetium
isotope impurities, namely 93Tc, 94Tc, 95Tc, 95mTc, 96Tc, and 97mTc, were
detected in the final formulation. The evaluated radionuclidic purity was
99.78% at EOS and remained greater than 99.0% up to 24 h from EOS. No
isotopes of Mo and Nb were found in the final formulation: 92mNb, 96Nb,
97Nb, 99Mo, 99mRh, and 100Rh were quantitatively extracted in the waste
fraction during the purification procedure.
Conclusions: We showed that the radionuclidic purity of 99mTc pro-
duced with a PET cyclotron at 16.5 MeV is fully adequate to clinical use,
provided that the isotopic composition of the starting molybdenum and
irradiation parameters (energy, time) are appropriately selected.
C.347
PRELIMINARY RADIONUCLIDIC PURITY ASSESSMENT OF 223RA
A. Corazza V. Trojan G. Cicoria F. Zagni D. Pancaldi M. Marengo
a Postgraduate School in Medical Physics, University of Bologna, Bologna, Italy;
b Physics Department, University of Bologna, Bologna, Italy; c Medical Physics
Department, University Hospital S.Orsola – Malpighi, Bologna, Italy
Introduction: 223Ra-dichloride (t1/2 = 11.43 d) is produced by a 227Ac/
227Th generator having a t1/2 of 21.8 years. The final product could possibly
contain contaminants as 227Th, due to the breakthrough of the parent, or
226Ra (1600 y) and 228Th (1.9 y), due to the production of 227Ac. The aim
of this study is to analyze different samples of 223Ra-dichloride at differ-
ent times after the reference date, in order to assess purity.
Material and Methods: 13 samples from different production batches of
223Ra were prepared taking 1 mL of solution from the original batch (con-
centration of 1 MBq/mL). Reference time varies from 5 to 16 months before
our measurements. Each sample was analyzed using a calibrated HPGe
gamma spectrometer (efficiency 30%, resolution 1.8 keV at 1332 keV), using
the same geometry for all samples.
In the most recent samples 227Ac presence was assessed considering only
the emission of 227Th, while in the other samples 223Ra presence was also
considered. The presence of 226Ra, 228Th and their decay product was
evaluated for all the samples prepared.
Activity of each radionuclide, Minimum Detectable Activity (MDA) and back-
ground subtraction were computed using GENIE 2000 software.
Results: In all but one sample no impurities were found and radionuclidic
purity was evaluated taking their activity equal to the MDA. The mean MDA
of 227Th, 226Ra and 228Th were 4 Bq/g, 1.8 Bq/g, and 1.7 Bq/g.
One single sample showed the presence of 227Ac, with a concentration of
14.1 ± 0.7 Bq/g. This sample is now under further investigation.
Mean radionuclidic purity is greater than 99.998%.
Conclusions: 223Ra-dichloride showed a very high radionuclidic purity.
This result guarantees that no unnecessary dose is given to the patient
because of impurities. Moreover the absence of other radionuclides sim-
plifies licensing procedures and waste management.
C.348
PERFORMANCE EVALUATION OF AN ADAPTIVE THRESHOLD
SEGMENTATION SOFTWARE FOR PET LESION DELINEATION
D. D’Ambrosio E. Antonioli L. Alberizzi C. Vellani D. Fantinato
a Servizio di Fisica Sanitaria, Fondazione S. Maugeri, Pavia, Italy; b Dipartimento
di Fisica, Università degli studi di Pavia, Pavia, Italy; c Servizio di Medicina
Nucleare, Fondazione S. Maugeri, Pavia, Italy
Introduction: The aim of this study was to assess the performance of adap-
tive thresholding algorithms (ATA) implemented in iTaRT SW (TA s.r.l.).
Materials and Methods: List mode PET data of NEMA IQ phantom (spheres
internal diameters ID = 10–37 mm) filled with 18F-FDG (sphere/background
SBR= 2 and 4) were acquired using Discovery690 PET/CT scanner (D690,
GEHC) and reorganized to get different scan duration (SD = 5, 15, 30, 45 min).
Data were reconstructed with OSEM (2, 3, 4 it; 6, 12, 18, 24 sub; 3, 6, 9 mm
Gaussian smoothing filter; standard axial filter) with (VPFXS) and without
(VPHD) TOF and PSF. The repeatability was evaluated performing 8 mea-
surements on the images with SBR = 4, SD = 5, 15 min and VPFXS
reconstruction by using 2D and 3D ATA of iTaRT SW. Coefficients of vari-
ation (CV) were calculated. For each set of OSEM parameters and SD, spheres
were segmented. The influence of SD and OSEM parameters was evalu-
ated by means of CV. ATA accuracy was estimated as percentage difference
of estimated spheres radius with respect to the measured CT one.
Results: Repeatability-High CV were found for small spheres (ID = 10, 13mm)
for both SD and different OSEM parameters up to 46.4% and 37.0% for 2D
and 3D ATA, respectively. Smaller CV were found up to 15.0%, 9.0%, 5.1%
and 4.7% for 2D ATA and 29.2%, 8.6%, 8.0% and 6.6% for 3D ATA, respec-
tively for ID = 17, 22, 28, 37 mm. SD and OSEM parameters influence-For
sphere ID = 22, 28, 37 mm, varying SD led to CV less than 6.3% (VPFXS) and
11.2% (VPHD) for 3D ATA and less than 9.5% for 2D ATA (VPHD and VPFXS).
Accuracy-For SBR = 4, accuracy was less than 1.8% (VPFXS) and 3.7% (VPHD)
for 3D ATA and less than 5.7% (VPFXS) and 9.3% (VPHD) for 2D ATA. Results
are relative to spheres ID= 22–37 mm. Accuracy increased up to 56.4% for
smaller volumes. Spheres contouring sometimes failed for SBR = 2 and higher
accuracy values were estimated.
Conclusions: Performance evaluation of ATA applied to D690 PET images
was performed as it is important before use in clinical practice.
C.349
QUANTITATIVE SPECT/CT IMAGING USING A COMMERCIAL SOFTWARE
L. D’Ambrosio L. Aloj P. Chiaramid V. Cerciell P. Gaball A. Prisc
S. Lastori a Istituto Nazionale Tumori, IRCCS Fondazione Pascale, Napoli, Italy;
b GE Healthcare, Roma, Italy
Introduction: Routine SPECT imaging lacks quantitative information. The
use of CT attenuation correction and lesion localization based on the fusion
of diagnostic CT and SPECT datasets can improve accuracy. This work aims
to validate commercially available software (Q.Metrix, GE Healthcare) that
provides absolute quantification in order to apply this software in clinical
practice.
Materials and Methods: For absolute quantification Q.Metrix needs data
of patients, injected activity information and camera sensitivity. This soft-
ware includes tools for reconstructing SPECT/CT data with attenuation,
scatter and collimator depth-dependent three-dimensional resolution re-
covery correction. Furthermore, it includes also segmentation tools both
on CT and NM images. All acquisitions were performed using hybrid dual-
head SPECT-CT camera (Discovery 670, GE Healthcare). Camera sensitivity
was evaluated for Tc99m and I131 both with Petri dish and point source.
Accuracy of activity quantification was performed on a large homoge-
neous source with addition of attenuating/scattering medium. A NEMA/
IEC body phantom was utilized to obtain recovery coefficient curve.
e102 Abstracts/Physica Medica 32 (2016) e97–e115