A system for patient set-up in external beam radiotherapy was developed using Augmented Reality (AR). Live images of the linac treatment couch and patient were obtained with video cameras and displayed on a nearby monitor. A 3D model of the patient's external contour was obtained from planning CT data, and AR tracking software was used to superimpose the model onto the video images in the correct position for treatment. Throughout set-up and treatment, the user can view the monitor and visually confirm that the patient is positioned correctly. To ensure that the virtual contour was displayed in the correct position, a process was devised to register the coordinates of the linac with the camera images. A cube with AR tracking markers attached to its faces was constructed for alignment with the isocentre using room lasers or cone-beam CT. The performance of the system was investigated in a clinical environment by using it to position an anthropomorphic phantom without the aid of additional set-up methods. The positioning errors were determined by means of CBCT and image registration. The translational set-up errors were found to be less than 2.4 mm and the rotational errors less than 0.3 degrees. This proof-of-principle study has demonstrated the feasibility of using AR for patient position and pose guidance.

No abstract