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

114 / 146

precision in quantitative 18FDG oncological studies for monitoring re-
sponse to treatment.
C.370
A NEW REGULARIZED RECONSTRUCTION ALGORITHM IN THE
EVALUATION OF CONTRAST DETECTABILITY AND BACKGROUND
VARIABILITY: COMPARISON WITH STANDARD OSEM ALGORITHM
S. Morzenti *, A. Zorz, E. De Ponti, A. Crespi. A.O. San Gerardo, S.C. Fisica
Sanitaria, Monza, Italy
Introduction: The GE Discovery IQ PET/CT scanner is equipped with the
new Q.Clear regularized reconstruction iterative algorithm (RR); it incor-
porates the resolution recovery option and setting only one parameter (b-
value) is always able to reach the full convergence (25 iterations) of the
data without increasing the image noise.
Material and Methods: The IEC body phantom, filled with 52 MBq of
18F-FDG according to NEMA_NU2-2012 protocol, was acquired with two
different spheres to background (S/B) concentration ratio (4:1, 8:1). Data
sets were reconstructed using three iterative algorithms: the standard
OSEM (VPHD), OSEM plus resolution recovery (VPHD-S) and RR one
(Q.Clear).
Contrast Detectability (CD) and Background Variability (BV), calculated
according to NEMA definition, were analyzed as a function of spheres
diameter (from 10 to 37 mm) and reconstruction parameters (number of
iterations from 3 to 25 for VPHD and VPHD-S and b-value from 5 to 150
for Q.Clear).
Results: CD increases with the number of iterations in both VPHD and
VPHD-S and, established the iteration, applying the resolution recovery
option; using the Q.Clear algorithm the CD of the two smallest spheres de-
crease increasing the b-value, but it remains constant for the other ones
in the range 0–50. For all the diameters the BV decreases from VPHD to
VPHD-S using less than 10 iterations; the opposite effect is obtained with
more iterations especially in the bigger diameters. Using Q.Clear the BV de-
creases increasing the b-factor value independently of the ROI dimension;
for b-value greater than 50 this effect increases for the two smallest ROIs.
The same results are obtained for both the S/B ratio.
Conclusion: The contrast values obtained with Q.Clear and VPHD-S with
25 iterations are comparable, but changing only the b-value it is possible
to optimize the image quality of the first reconstruction; in particular, in
IEC phantom, increasing the b-value up to 50 the noise is reduced keeping
stable the contrast.
C.371
NOVEL LABR3:CE GAMMA CAMERA PROTOTYPE FOR ACCURATE
COMPTON SCATTER REJECTION
C. Orlandi M.N. Cint R. Pellegrini R. Pani a Department of Molecular
Medicine, Sapienza University of Rome, Rome, Italy; b Enterprise Risk
Management/Medical Physics, Bambino Gesù Children’s Hospital, IRCCS, Rome,
Italy; c Medico-Surgical Sciences and Biotechnologies, Sapienza University of
Rome, Rome, Italy
Introduction: The improvement of Nuclear Medicine imaging is strongly
dependent on the advances of spatial and energy resolution. Clinical gamma
cameras are limited by fixed geometry that is not adequate for spatial res-
olution enhancement, and by the characteristics of typically used scintillation
crystal (NaI:Tl). A novel detector prototype is proposed with the aim to
achieve outstanding energy resolution and optimal spatial resolution per-
formances, compared to clinical gamma camera ones.
Materials and Methods: The presented prototype consists of a small field
of view (FOV) gamma camera based on a Lanthanum Tri-Bromide (LaBr3:Ce)
scintillation crystal with round shape and reflective treatment of sur-
faces, optically coupled with a multi-anode photomultiplier tube. The
outstanding performances expected for the presented prototype are veri-
fied by a complete characterization in terms of intrinsic parameters (energy
resolution, intrinsic spatial resolution, position linearity). In addition, mea-
surements of activity ratios are performed by using Picker Nuclear Thyroid
phantom.
Results: The prototype shows spatial resolution of 2 mm and 7.5% energy
resolution at 140 keV. This result is principally due to the high light yield
of the scintillator and to the surface treatment of the crystal. The high energy
resolution of LaBr3:Ce gamma camera ensures an enhancement in image
contrast, thanks to the more accurate Compton rejection. LaBr3:Ce device
is able to evaluate the activity ratios in cold and hot phantom regions from
images corresponding to different energy windows more precisely than with
clinical gamma cameras.
Conclusion: The results demonstrate the real effectiveness of the enhance-
ment of spatial and energy resolutions on the improvements of Nuclear
Medicine imaging. The prototype could be used for quantitative imaging
in metabolic radiotherapy, where its characteristics would guarantee a more
accurate evaluation of image counts, and thus of absorbed dose.
C.372
A MULTICENTER DOSIMETRIC AND OBSERVATIONAL STUDY FOR LESION
DOSIMETRY IN 223RA THERAPY OF BONE METASTASES: CALIBRATION
PROTOCOL AND ELIGIBILITY CRITERIA
M. Pacilio * B. Cassano C. Chiesa M. Ferrari C. Pettinat E. Amato
F. Fioroni L. Lorenzon R. Pellegrini E. Di Castro R. Pani
M. Cremones a Department of Medical Physics, Azienda Ospedaliera San
Camillo Forlanini, Roma, Italy; b Postgraduate School of Medical Physics,
Sapienza University of Rome, Roma, Italy; c Department of Nuclear Medicine,
Istituto Nazionale Tumori IRCCS Foundation, Milano, Italy; d Department of
Medical Physics, Istituto Europeo di Oncologia, Milano, Italy; e Department of
Medical Physics, Policlinico S.Orsola-Malpighi, Bologna, Italy; f Section of
Radiological Sciences, Department of Biomedical and Dental Sciences and of
Morphologic and Functional Imaging, Messina, Italy; g Department of Medical
Physics, IRCCS S. Maria Nuova, Reggio Emilia, Italy; h Department of Molecular
Medicine, Sapienza University of Rome, Roma, Italy; i Department of
Radiological, Oncological and Anatomo Pathological Sciences, Sapienza
University of Rome, Roma, Italy
Introduction: Multicenter studies could help standardize the dosimetric
approach and investigate the role of lesion dosimetry for 223Ra therapy.
This work was aimed to investigate gamma camera calibration protocols
and patient eligibility criteria for dosimetric studies with conjugated view
technique.
Materials and Methods: Calibrations with 223Ra were performed with
two gamma cameras (3/8-inch crystal) acquiring planar static images with
double-peak (82 and 154 keV, 20% wide) and MEGP collimator. The sen-
sitivity was measured in air by varying activity, source–detector distance,
and source (i.e., Petri dishes) diameter. Transmission curves were mea-
sured for attenuation and scatter correction with the pseudoextrapolation
number method. 39 lesions (14 patients) were studied after 223Ra
administration with three planar images of about 30 min at 1, 2–3 and
7–15 days. The visibility on 223Ra images of lesions detected from an-
teroposterior 99mTc-MDP WB images was studied by ROC curve, using
the lesion 99mTc contrast ratio with respect to surrounding soft tissue as
score value.
Results: Sensitivity was nearly constant varying activity and distance, for
source diameter higher than 3.5 cm. Main eligibility criteria for dosimet-
ric studies resulted: 1. patient compliance, 2. lesions not overlapped to
gastrointestinal (GI) tract (due to high 223Ra uptake), 3. visibility of lesions
on 223Ra images. The ROC analysis yielded an AUC of 0.852 and an optimal
discriminant threshold of 10, corresponding to a maximum accuracy of about
86%.
Conclusion: The minimum calibration protocol requires sensitivity and
transmission curve measurements, but recovery coefficients would be also
needed for small lesions. Lesions with 99mTc contrast ratio higher than
10 not overlapped to the GI tract are generally visible on 223Ra images,
and thus eligible for dosimetric studies, but lesion delineation on 99mTc-
MDP WB images and WB/static image coregistration for accurate ROI
overlapping are mandatory.
e109 Abstracts/Physica Medica 32 (2016) e97–e115