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

62 / 146

relative to PFD, with a slight overestimation of the tails (<1%), due to the
absence of the shielding. Output factors were in good agreement for fields
up to 5 × 5 cm2 (<1%).
Conclusions: The Razor diode has the same spatial resolution of SFD with
the additional advantages of improved stability, becoming a possible can-
didate for small field dosimetry.
A.192
A PRELIMINARY DOSIMETRIC STUDY TO VALIDATE MULTI-SEGMENT PLAN
(FIELD IN FIELD) FOR BREAST IRRADIATION USING IN VIVO DOSIMETRY
(IVD)
S. Riccardi *, R. Nigro, R. Maglio, M.G. Mangiacotti, R. Cassese. S. Camillo de
Lellis Hospital, Rieti, Italy
Purpose: IVD represents an independent check of the dose able to high-
light the reproducibility of the treatment over time and able to help avoiding
systematic errors during dose delivery. Moreover, IVD may be useful to val-
idate a new technique. The aim of this study is to evaluate the accuracy
and reproducibility of dose distributions of breast treatments using field-
in-field (FIF) technique (TPS Eclipse-Varian) with SOFTDISO (Best Medical
Italy Srl). The FIF technique is a forward planning approach that uses multi-
leaf collimator sub-fields to improve dose uniformity throughout the whole
breast volume.
Method and materials: We acquired the portal image during each session
for all beams of seven breast FIF treatment planned. The treatments were
delivered with 6 MV photon beam by a Varian Clinac 2100 C/D. CT images,
RT plan and portal images of patients were imported in SOFTDISO. SOFTDISO
is provided with an analysis module dedicated to the reconstruction of the
dose at the point isocenter from the portal images signal. The value of the
transit signal is converted to a dose value, Diso, and compared with the
value of calculated dose, Diso,TPS. The comparison is performed through
the index R = Diso/Diso,TPS. The acceptance criterion was: 0.95 ≤ R ≤ 1.05.
The daily EPID images were compared to a reference image obtained in the
first session of therapy. A further module is dedicated to the gamma-
analysis between the reference and daily portal image.
Results: The comparisons showed good results. The R values for 5 treat-
ments were within 1.5%. In two treatments were within 4%. In particular
the R average for all sessions valued was 1.016 ± 0.021. The gamma-
analysis, performed with 3 mm/5% criteria, yielded Pg < 1 ≥ 93% in 80% of
the tests, 20% of the tests supplied 88% ≤ Pg < 1 < 93% due to small setup
variations.
Conclusions: The IVD assures that the FIF treatment has a good reproduc-
ibility during treatment sessions and can replace the standard breast
treatment with dynamic wedges.
A.193
INTRAFRACTION BREAST DISPLACEMENTS DURING HELICAL
TOMOTHERAPY
R. Ricotti D. Ciardo G. Fattor M.C. Leonardi A. Morra F. Pansin
R. Cambri F. Cattani C. Spinell C. Gianol M. Riboldi G. Baroni
B.A. Jereczek-Foss R. Orecchia a Department of Radiation Oncology,
European Institute of Oncology, Milan, Italy; b Dipartimento di Elettronica
Informazione e Bioingegneria, Politecnico di Milano, Milan, Italy; c Paul Scherrer
Institut, Villigen, Switzerland; d Unit of Medical Physics, European Institute of
Oncology, Milan, Italy; e Tecniche di Radiologia Medica, per Immagini e
Radioterapia, University of Milan, Milan, Italy; f Ludwig Maximilians University,
Munich, Germany; g Bioengineering Unit, Centro Nazionale di Adroterapia
Oncologica, Pavia, Italy; h Department of Oncology and Hemato-oncology,
University of Milan, Milan, Italy
Purpose: To investigate intra-fraction breast motion during long-lasting (10–
20 min) breast helical tomotherapy by means of optical tracking.
Materials and methods: Twenty locoregional breast cancer patients un-
derwent helical tomotherapy irradiation after receiving conservative surgery
or mastectomy. Non-invasive monitoring of respiratory motion was achieved
through infrared tracking of a passive marker placed near the surgical scar.
In order to obtain the displacement deriving from the patient movement
only, we subtracted the trace of an additional marker placed on the treat-
ment couch. Respiratory signals were analyzed in terms of peak-to-peak
amplitudes and baseline drifts, obtained by low-pass moving average fil-
tering with a time window of 60 sec.
Anisotropic clinical target volume (CTV) safety margin expansion due to
the measured intra-fraction motion was calculated relying on a synthetic
representation of the specific patient respiratory pattern, obtained by adding
half of the most probable respiratory amplitude to the non-respiratory move-
ment of the scar trace.
Result: The most probable measured breathing amplitude was less than
2 mm along all anatomical directions.
Intra-fraction baseline drifts were mostly in the posterior and inferior di-
rections for all patients in most fractions, with the exception of one patient,
who exhibited a countertendency behavior. The distribution of right–left
shifts resulted in almost zero median, with a narrow interquartile range.
Resulting anisotropic safety margin expansions across all patients with the
exception of the outlier patient were 1.58–2.44 mm in right–left, 4.41–
3.65 mm in inferior–superior and 3.78–2.15 mm in the posterior–anterior
directions, respectively.
Conclusion: Non-respiratory motion during prolonged treatment induces
notable uncertainties. Non-invasive continuous monitoring of patient set-
up variations including baseline drifts is considered necessary in order to
minimize dosimetric deviations.
A.194
DOSIMETRIC PROPERTIES OF 3D-PRINTED FLAT BOLUS FOR EXTERNAL
RADIOTHERAPY
R. Ricotti A. Vavassor R. Spoto D. Ciard F. Pansini A. Bazani
S. Noris F. Cattani R. Orecchia B.A. Jereczek-Fossa a Department of
Radiation Oncology, European Institute of Oncology, Milano, Italy; b Department
of Oncology and Hemato-oncology, University of Milan, Milano, Italy; c Unit
of Medical Physics, European Institute of Oncology, Milano, Italy; d Tecniche
di Radiologia Medica, per Immagini e Radioterapia, University of Milan, Milano,
Italy
Purpose: Dosimetric property evaluation of acrylonitrile butadiene styrene
(ABS) and polylactide (PLA) plastics, and their suitability for bolus 3D print-
ing applied in high-energy radiotherapy to overcome the skin-sparing effect.
Materials and methods: Eight flat boluses for the water-equivalent RW3
slab phantom were 3D printed using a commercial 3D printer (HP3DX100,
Hamlet, Dublin, IE) using ABS and PLA. Flat boluses were sized at 8 × 8 × 1
cm (length × width × thickness). Each bolus was printed with different infill
percentages (10%, 20% 40% and 60%). The mean value of Hounsfield unit
(HU) of the 3D printed boluses was provided by analyzing their com-
puted tomography (CT) scans.
The measurements were performed with Vero® System (Brainlab AG,
Feldkirchen, Germany) delivering 200 monitor units with a 6 MV photon
beam, 5 × 5 cm open field with 90-degree gantry angle at 100 cm surface
to surface distance on a water-equivalent phantom in three configura-
tions: without bolus, with a commercial bolus and with the eight 3D printed
boluses. Gafchromic EBT3 film (International Specialty Products, Wayne,
NJ) placed between phantom slabs provided dose profile measurements.
Results: Negative HU were due to the air gap inside the flat bolus infill
pattern. The mean HU increased with the percentage infill, resulting in higher
bolus density in a reduced distance from the surface of the phantom where
the maximum dose occurs. Build-up peaks shifted toward the phantom
surface when any bolus was used. ABS and PLA boluses with an infill per-
centage of 40% had comparable performance to the commercial bolus.
Conclusion: 3D printed flat boluses can decrease the skin-sparing effect
as a commercially available bolus. The performed analysis accurately de-
scribes the physical behavior of these plastic materials. Patient-specific
boluses could be outlined from patient CT images and 3D printed shaping
the actual anatomy of the patient and improving the fitting between bolus
and skin surfaces.
e57 Abstracts/Physica Medica 32 (2016) e1–e70