Treatment Planning for a Radiosurgical System with General Kinematics

with General Kinematics Achim Schweikard ?y Rhea Tombropoulos ?z Lydia Kavraki ? John R. Adler y Jean-Claude Latombe ? ? Robotics Laboratory, Department of Computer Science y Department of Neurosurgery z Section on Medical Informatics, School of Medicine Stanford University, Stanford, CA 94305, USA, [email protected] Abstract: In radiosurgery a beam of radiation is used as an ablative surgical instrument to destroy brain tumors. Treatment planning consists of computing a sequence of beam con gurations for delivering a necrotic dose to the tumor, without damaging healthy tissue or particularly critical structures. In current systems, kinematic limitations severely constrain beam motion. This often results in inappropriate dose distributions. A new radiosurgical system has been implemented to overcome this disadvantage. In this system, a compact radiation source of high energy is moved by a 6-dof robotic arm. We describe algorithms for computing a motion with speci ed characteristics for this new system. Treatment plans used at test sites with earlier systems are compared to plans computed with the described algorithms. The experience reported shows that full kinematic exibility combined with treatment planning algorithms allows for better protection of healthy tissue and higher dosage in tumors. Acknowledgments: This research is funded in part by the Sheik Enany Fund, Lorraine Ulshafer Fund and by the National Library of Medicine (LM-05305 and LM-07033). The authors thank Mohan Bodduluri, Joe Depp, and Bo Preising of Accuray Inc. for their help with the Neurotron 1000.