Development of a phantom to evaluate the positioning accuracy of patient immobilization systems using thermoplastic mask and polyurethane cradle.

PURPOSE: The purpose of this study was to develop a new phantom to evaluate the positioning accuracy of patient immobilization systems. METHODS: The phantom was made of papers formed into a human shape, paper clay, and filling rigid polyester. Acrylonitrile butadiene styrene (ABS) pipes were inserted at anterior-posterior (A-P) and right-left (R-L) directions in the phantom to give static load by pulling ropes through the pipes. First, the positioning precision of the phantom utilizing a target locating system (TLS) was evaluated by moving the phantom on a couch along inferior-superior (I-S), A-P, and R-L directions in a range from -5 mm to +5 mm. The phantom's positions detected with the TLS were compared with values measured by a vernier caliper. Second, the phantom movements in a tensile test were chosen from patient movements determined from 15 patients treated for intracranial lesions and immobilized with a thermoplastic mask and polyurethane cradle. The phantom movement was given by minimum or maximum values of patient movements in each direction. Finally, the relationship between phantom movements and the static load in the tensile test was characterized from measurements using the new phantom and the TLS. RESULTS: The differences in all positions between the vernier caliper measurement and the TLS detected values were within 0.2 mm with frequencies of 100%, 95%, and 90% in I-S, A-P, and R-L directions, respectively. The phantom movements according to patient movements in clinical application in I-S, A-P, and R-L directions were within 0.58 mm, 0.94 mm, and 0.93 mm from the mean value plus standard deviation, respectively. The regression lines between the phantom movements and static load were given by y = 0.359x, y = 0.241x, and y = 0.451x in I-S, A-P, and R-L directions, respectively, where x is the phantom movement (mm) and y is the static load (kgf). The relationship between the phantom movements and static load may represent the performance of inhibiting patient movements, so the accuracy of the immobilization system in the intracranial lesion will be estimated in advance by basic tensile test on the new phantom. CONCLUSIONS: The newly developed phantom was useful to evaluate the accuracy of immobilization systems for a Cyberknife system for intracranial lesions.

Influence of collimator size on three-dimensional conformal radiotherapy of the cyberknife.

PURPOSE: In stereotactic conformal radiotherapy of the CyberKnife, multiple narrow beams from a cylindrical collimator are delivered to a planning target volume (PTV) without an isocenter by inverse planning. The influence of collimator size on stereotactic conformal radiotherapy was examined. MATERIALS AND METHODS: Five targets including two spherical targets (2.1 ml, 16.4 ml), an ellipsoidal target (2.9 ml), and two irregularly shaped targets (1.6 ml, 6.6 ml) were evaluated. Conformity and homogeneity of dose distribution, doses covering 95% volume of PTVs (D95), dose volume histograms (DVHs), and patterns of dose distribution were investigated in relation to collimator size for each target. We evaluated conformity and homogeneity using VD95/PTV (VD95 = volume covered by a D95 isodose surface) and Dmax/Dmin (Dmax, Dmin = maximum and minimum dose within the PTV, respectively), respectively. RESULTS: Conformity and homogeneity were not significantly influenced by collimator size. By using a considerably smaller collimator compared with the minor axis of the target the central dose of the target fell by less than 10% and D95 often decreased slightly. CONCLUSION: In stereotactic conformal radiotherapy of the CyberKnife, a smaller collimator tended to reduce the central dose and D95. These characteristics of the CyberKnife system should be considered.