ABSTRACT
PURPOSE: Positioning verification is usually performed with treatment beam (MV) portal images (PI) using an electronic portal imaging device (EPID). A new alternative is the use of a low energy photon source (kV) and an additional EPID mounted to the accelerator gantry. This system may be used for PI or--with rotating gantry--as cone-beam CT (CBCT). The dose delivered to the patient by different imaging processes was measured. METHODS AND MATERIALS: A total of 15 in-vivo dose measurements were done in five patients receiving prostate IMRT. For anterior-posterior (AP) and lateral PI with MV and kV photons measurement points were inside the rectum and at the patient's skin. Dose for CBCT was measured in the rectum. Additional measurements for CBCT were done in a cylindrical CT-dose-index (CTDI) phantom to determine peripheral, central and weighted CTDI. RESULTS: The dose for AP MV PI was 57.8 mGy at the surface and 33.9 mGy in the rectum, for lateral MV PI 69.4 mGy and 31.7 mGy, respectively (5 MU/exposure). The dose for AP kV PI was 0.8 mGy at the surface and 0.2 mGy in the rectum, for lateral PI 1.1 mGy and 0. 1 mGy, respectively. For a CBCT the rectal dose was 17.2 mGy. The peripheral CTDI was 23.6 mGy and the center dose was 10.2 mGy, resulting in a weighted CTDI of 19.1 mGy in the phantom and an estimated surface dose of < or =28 mGy. CONCLUSIONS: Even taking into account an RBE (Relative Biological Effectiveness) of 2 for kV vs. MV radiation, for kV PI the delivered dose is lower and image quality is better than for MV PI. CBCT provides a 3D-image dataset and dose exposure for one scan is lower than for two MV PI, thus rendering frequent volume imaging during a fractionated course of radiotherapy possible.
Subject(s)
Cone-Beam Computed Tomography/methods , Radiotherapy Dosage , Radiotherapy Planning, Computer-Assisted/methods , Humans , Male , Phantoms, Imaging , Prostatic Neoplasms/radiotherapy , Relative Biological EffectivenessABSTRACT
PURPOSE: The repositioning accuracy of mask-based fixation systems has been assessed with two-dimensional/two-dimensional or two-dimensional/three-dimensional (3D) matching. We analyzed the accuracy of commercially available head mask systems, using true 3D/3D matching, with X-ray volume imaging and cone-beam CT. METHODS AND MATERIALS: Twenty-one patients receiving radiotherapy (intracranial/head-and-neck tumors) were evaluated (14 patients with rigid and 7 with thermoplastic masks). X-ray volume imaging was analyzed online and offline separately for the skull and neck regions. Translation/rotation errors of the target isocenter were analyzed. Four patients were treated to neck sites. For these patients, repositioning was aided by additional body tattoos. A separate analysis of the setup error on the basis of the registration of the cervical vertebra was performed. The residual error after correction and intrafractional motility were calculated. RESULTS: The mean length of the displacement vector for rigid masks was 0.312 +/- 0.152 cm (intracranial) and 0.586 +/- 0.294 cm (neck). For the thermoplastic masks, the value was 0.472 +/- 0.174 cm (intracranial) and 0.726 +/- 0.445 cm (neck). Rigid masks with body tattoos had a displacement vector length in the neck region of 0.35 +/- 0.197 cm. The intracranial residual error and intrafractional motility after X-ray volume imaging correction for rigid masks was 0.188 +/- 0.074 cm, and was 0.134 +/- 0.14 cm for thermoplastic masks. CONCLUSIONS: The results of our study have demonstrated that rigid masks have a high intracranial repositioning accuracy per se. Given the small residual error and intrafractional movement, thermoplastic masks may also be used for high-precision treatments when combined with cone-beam CT. The neck region repositioning accuracy was worse than the intracranial accuracy in both cases. However, body tattoos and image guidance improved the accuracy. Finally, the combination of both mask systems with 3D image guidance has the potential to replace therapy simulation and intracranial stereotaxy.
