Imaging Dose Dependence On Fractionation Scheme and Tracking Method in CyberKnife Robotic Radiosurgery
CJ Tien1*, S Dieterich2, S Lee1, BH Curran1, ES Sternick1, (1) Rhode Island Hospital/Brown Medical Center, Providence, RI, (2) UC Davis Medical Center, Sacramento, CA,SU-E-J-24 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: Imaging dose in robotic radiosurgery is analogous to cone-beam computed tomography imaging for stereotactic radiosurgery: there needs to be a balance between increasing imaging to increase targeting accuracy versus reducing imaging to lower imaging dose. We study the effect of fractionation scheme and tracking method upon reducing imaging dose.
Methods: The imaging system of the Accuray CyberKnife consists of two ceiling-mounted kilovoltage x-ray tubes and two accompanying floor-mounted amorphous silicon flat-panel detectors. The x-ray tubes have 2.5 mm Al filtration and are capable of 40-125 kV, 25-300 mA, and 1-500 ms. A cohort of 427 patients from two institutions using CyberKnife was retrospectively reviewed to determine the number of images taken during treatment. This database was analyzed to determine if different fractionation schemes or tracking methods affected the number of images.
Results: Over all patients, the median number of images taken was approximately 400, which was an estimated entrance skin exposure of about 0.5 Gy. This number was found to be most dependent upon the fractionation scheme and tracking method. The amount of imaging increased 400% when switching between one-fraction and three-fraction treatments. Additionally, the amount of imaging increased 150% between skull and spinal tracking. Between the two institutions, differences were found in the absolute magnitude in images taken due to the sampling interval chosen. Dose differences were also observed as a result of operator adjustment of kV, mAs, or exposure time.
Conclusion: This study provides supplemental information to AAPM TG-75, which did not examine imaging dose in robotic radiosurgery. The imaging dose was found to vary widely but was most strongly dependent upon fractionation scheme and tracking method. We identified three methods with the potential to decrease imaging dose by 300%.