PurposeNon-coplanar beams and arcs are routinely used to improve dosimetry for intracranial cases but their application for extra-cranial cases has been hampered by the risk of collision. This has led to conservative beam selection whose impact on plan dosimetry has not been previously studied. Methods and materialA full-body 3D patient surface was acquired using optical cameras for a single lung patient at the time of CT simulation. Eight SBRT plans were created for the patient, with varying degrees of non-coplanarity and deliverability. The plans included VMAT and IMRT plans ranging from simple, coplanar arcs to multiple non-coplanar arcs and IMRT beams. There was a total of 70 fields created across the 8 plans, of which 21 fields were undeliverable with a 5cm buffer. OAR metrics including R50, Dmax 2 cm from the PTV, lung V20 and chest wall V30 were evaluated. Five expert SBRT dosimetrists from five institutions evaluated field deliverability, with or without the guidance of the clearance map. ResultsIn the dosimetry evaluation, a clear trend in increasing dosimetric compactness and OAR sparing is observed with increasing plan non-coplanarity. R50, Dmax 2cm, lung V20, and chest wall V30 decreased 41%, 39%, 43%, 57%, respectively, from plan 1 (two coplanar partial arcs) to plan 8 (19 non-coplanar IMRT beams). In the observer tests, the expert dosimetrists’ ability to accurately discern beam deliverability due to collision significantly increases with the clearance map. The errors in predicting colliding fields were eliminated using the whole-body surface and clearance map, and the user was able to select fields based on plan quality and patient comfort instead of being overly conservative. ConclusionThe study shows that incorporating a personalized, whole-body clearance map in the treatment planning workflow can facilitate the adoption of non-coplanar beams or arcs that benefit the SBRT plan dosimetry.