Integration Of Optical Tracking With Dynamic Treatment Couch Real-Time Tracking For Intra-Fraction Motion Compensation

To integrate an optical tracking camera able to localize passive, infrared-reflective markers into a dynamic treatment couch-based tracking system, and to evaluate the performance of the motion compensating treatment couch after integration with the optical tracking camera on subjects undergoing normal respiration. Marker localization latency and overall system tracking accuracy were evaluated. Marker localization latency comprises camera system processing time and delays in communicating the marker position to the system's motion controller. To measure the localization latency, a marker was affixed to the couch, which was then directed to move through a sinusoidal path in each of three axes. Cross-correlation was used to calculate the delay between the tracking camera localizations and the real-time analog encoder feedback of the motors. To evaluate the tracking accuracy of the system as a whole, the residual motion of a marker during real-time respiratory motion compensation was measured in 5 volunteer subjects. Subjects were positioned on the couch in supine position while a marker affixed to the skin halfway between the xyphoid and the umbilicus was tracked with the camera. The tracking signal was used to direct real-time couch motion such that the marker was stationary with respect to the treatment room in the superior-inferior (SI) direction. The motion of the marker without couch compensation and the residual motion of the marker after couch compensation were analyzed for 5 min of free breathing from each subject. The localization latency associated with the camera was approximately 0.02 sec. The free-breathing SI end-exhale position varied by 2.4 mm to 4.2 mm in the four subjects. Peak-to-peak SI marker motion ranged from 6.1 mm to 15.2 mm before correction, with 32.1% of points more than 2 mm from the setup position. The mean ± SD of marker displacements with respect to the treatment room after couch-based motion compensation was 0.0 ± 0.5 mm, with 0.1% of points greater than 2 mm from the setup position. The tracking system compensated for the respiratory motion of the subjects' abdominal surfaces, as measured by the tracking camera. The distribution of marker displacements was centered about the setup position, and the SD of marker displacements was less than 1 mm.