Feasibility of using respiratory correlated mega voltage cone beam computed tomography to measure tumor motion.

The purpose of this study was to test the feasibility of using respiratory correlated mega voltage cone-beam computed tomography (MVCBCT), taken during patient localization, to quantify the size and motion of lung tumors. An imaging phantom was constructed of a basswood frame embedded with six different-sized spherical pieces of paraffin wax. The Quasar respiratory motion phantom was programmed to move the imaging phantom using typical respiratory motion. The moving imaging phantom was scanned using various MVCBCT imaging parameters, including two beam line types, two protocols with different ranges of rotation and different imaging doses. A static phantom was also imaged as a control. For all the 3D volumetric images, the contours of the six spherical inserts were measured manually. Compared with the nominal sphere diameter, the average relative error in the size of the respiratory correlated MVCBCT spheres ranged from 5.3% to 12.6% for the four largest spheres, ranging in size from 3.6 cc to 29 cc. Larger errors were recorded for the two smallest inserts. The average relative error in motion was 5.1% smaller than the programmed amplitude of 3.0 cm. We are able to conclude that it is feasible to use respiratory correlated MVCBCT to quantify tumor motion for lung cancer patients.