Michele Zeverino

ASL AT, Ospedale "Cardinal Massaia", ASTI A.S.O. San Giovanni Battista, Presidio Ospedaliero "San Giovanni Antica Sede", TORINO Relatori: Ugo Nastasi, Simonetta Amerio Introduction IMRT implementation definitively requires time efforts. Beside the characterization of the MLC, particular attention should be paid also to the performances of the linear accelerator in extreme situations, like as the machine output for small field sizes and for small MU deliveries. Moreover, for IMRT dosimetry, the proper choice of the detector is not always straightforward as it is for 3D-CRT dosimetry due to the needing to perform measurements in non equilibrium of charged particles conditions. In this thesis both machine output and MLC performances were accurately evaluated for a Varian Clinac 600DBX equipped with a Varian Millennium 120 multi-leaf collimator by means of different detectors and for different measurements settings. This was done in order to attain a cohomprensive workflow to adopt whenever an IMRT clinical implementation is requested. Results and discussion Small field dosimetry was carried out by use of two cylindrical ionization chambers different in their size. The performances of a Scanditronix Wellhofer CC13 (active voume 13 cc) and Scanditronix Wellhofer CC13 (active volume 4 cc) were compared in measurements of either dose profiles and output factors for field size ranging from 5 x 5 cm2 to 1 x 1 cm2. Furthermore, the same ionization chambers were used to measure PDDs for field sizes down to 2 x 2 cm2. Results suggested the use of the smaller detector between the two in exam in small field dosimetry. In fact, performances of CC13 and CC04 ionization chambers were comparable for fields sizes down to 2 x 2 cm2 in terms of dose profiles and output factors. Discrepancies were observed for the smallest field size measured, 1 x 1 cm2, since CC13 under estimated the output factor of about 12 % in respect to the CC04 reading. This is a consequence of the "dose blurring phenomenon" which took place whenever the dimensions of the detectors are similar to the dimensions of the field size causing an over estimation of the radiation field size. The difference in the size dimension of the two detectors lead also to difference in measured dose valued at shallow depths. Since the effective point of measure of the chamber is placed between the inner and the outer electrode, for shallow depths the two chambers remained with part of their total volume outside the medium resulting in a under estimation of the actual dose at that point. This effect decreases with depth and thus is more pronounced for larger detectors. CC13 always under estimated measured dose for the first 6 ­ 8 mm of depth in respect to the CC04 readings. The readings coming from the two chambers excellently matched afterwards. Although the just mentioned discrepancies observed between the two chambers may not be relevant for a 3D-CRT implementation, they assume significance in IMRT delivery suggesting the use of the smaller detector for small field dosimetry.

Linearity and stability of treatment fields delivered with small MU should be evaluated in particular for static IMRT delivery. A farmer-type ionization chamber was used for this purposes. Readings were collected for MU settings ranging from 1 to 100. Although the absorbed dose is linear with the corresponding MU delivered, results coming from stability measurements for small number of MUs revealed a significant over dosage. The charge collected for a single MU delivery was found to be greater than 8 % compared to one hundredth of the charge collected for 100 MU delivery. Such a discrepancy became smaller for increasing number of small MUs still remaining unacceptable for 5 MUs delivery (4 %). Such a phenomenon suggested to deliver clinical fields with a minimum number of MU greater than 5. Flatness and symmetry for small MU delivery were evaluated for MU values of 1, 2 , 3, 5, 10 and 15 by means of Gafchromic EBT films and Sunnuclear Mapcheck 2D diode array. Flatness resulting from film dosimetry suffered from the several source of uncertainties arising from the whole dosimetry process and thus calculated values were not taken as reliable. Flatness and symmetry resulting from diode array measurements as well as symmetry from film measurements were comparable to those determined for a standard 100 MU delivery, confirmed the capability of the linear accelerator to deliver small MU fields with an acceptable stability. Characterization of the MLC was the most time consuming process. Leaf positional accuracy as well leaf speed accuracy was checked by means of several detectors going from ionization chamber to radiochromic films passing through 2D diode array. Results carried out from different measurements always confirmed a perfect leaf accuracy in terms of positioning (better than 0.2 mm) and speed also independent from the orientation of both gantry and collimator. Rounded leaf end design lead mandatory to perform measurements in order to evaluate either the leaf offset in terms of variation between the radiation and the nominal field size and if the same offset is constant with the leaf position. The measured offset ranged from 0.9 to 1 mm for each leaf and it was observed to be constant in respect to the leaf positioning. Leaf side design produces the so-called "tongue and groove effect". If adjacent leaves are set to a certain position in respect to the isocenter in consecutive times they produce and under dosage in their overlapping area. The quantitative study of this effect was performed by means of radiochromic films and it was observed a dose deficiency ranging from 9 % to 12 % depending on the leaf number. In dynamic IMRT leaf pairs are not allowed to move closed. A gap of 0.5 mm (minimum leaf gap, MLG) is introduced by the MLC control system in order to avoid collisions between opposing leaves. As a consequence, radiation will pass through "closed" opposing leaves resulting in a small amount of dose delivered to the patient. The amount of dose due to the MLG was measured by radiochromic films and 2D diode array. Result showed that dose passing through the MLG is quantifiable in 5 % and 4 % in respect to an open field respectively for film and diode array dosimetry. The smaller percentage dose value resulted from diode reading was due to the size of the detector which was 0.3 mm larger than the MLG and thus slightly too wide to perform such measurements accurately.



2 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

Microsoft Word - Proof - 35 _pdf_.doc