Time delay study of a CT simulator in respiratory gated CT scanning.

In respiratory-gated radiotherapy (RGRT), if the time delay in a computed tomography (CT) simulator and that in a linear accelerator (Linac) are different, the simulation and the treatment cannot be synchronized. In this study, we presented a technique to measure the time delay of the AcQSim CT simulator (Philips Medical Systems, Cleveland, OH) using Varian's Real-Time Positioning Management (RPM) system (Varian Medical Systems, Palo Alto, CA). A respiratory gating platform (REF 91150, Standard Imaging, Inc., Middleton, MI) was first set at the position of amplitude maximum (phase 0). Then a ball of 1.3 cm diameter was put on the platform and set at the CT laser. A single axial scan was acquired across the center of the ball without motion. Then the motion was turned on and single axial scans gated at different phases were acquired with a very narrow gating window. The time between the phase giving a good estimate of the ball and phase 0 is the overall delay time. We found that for AcQSim CT, the overall delay for a single axial scan (with 1 s scan time) is 1.75 s. For multiple axial scans, the overall delay is 1.75 s for the first scan and 0.75 s for the subsequent ones. This demonstrated that the CT mechanical startup delay is 1 s. After the first axial scan, the overall delay per scan is less because CT gantry continuously spins and no mechanical delay exists. We call the overall delay without mechanical part the scanning delay, which basically equals half the scan time (0.5 s for 1 s scan time) plus the gating pulse triggering delay (250 ms). The delays were also verified using metal balls of 1.5 mm diameter set at the amplitude minimum (phase 180, initially). We note that it is the scanning delay rather than the triggering delay that should be compensated when doing motion-synchronized radiotherapy. The current interface between the RPM system and the AcQSim CT does not compensate for this scanning delay.