Program Information
Trajectory Optimization with Monte Carlo Tree Search Based AI for VMAT
P Dong*, H Liu , L Xing , Stanford Univ School of Medicine, Stanford, CA
Presentations
TU-D-108-6 (Tuesday, August 1, 2017) 11:00 AM - 12:15 PM Room: 108
Purpose: Artificial Intelligence (AI) recently beat the top GO players. With the help of AI, here we tried to tackle one of the most difficult optimization problem in radiation oncology: designing an efficient trajectory to deliver an isocentric VMAT treatment with an optimum dose distribution on a C-arm machine, where the gantry and couch move in coordination.
Methods: To search through the vast number of possible trajectories, we implemented Monte Carlo Tree Search (MCTS), the same method used by the AI which mastered the game of GO. Each trajectory picks the control points one by one. The next selected control point would need to be physically reachable and have the maximum Upper Confidence Bounds (UCT) for Trees, which is a function of average objective function value, the number of time this node has been selected and the total number of simulations. After forming a full trajectory, we run an inverse fluence map optimization with an infinity norm regularization term and the alternating direction method of multipliers, using all the control points on the trajectory. The obtained objective function value is then fed back to update the statistic of the control points on the trajectory.
Results: For both patient cases tested, the AI with MCTS found optimum trajectories within an hour after ~300 simulations with superior dosimetric qualities for OARs. For the skull case, the brain V30, V20 and V10 are reduced by 21%, 38% and 27% respectively. For the chest wall case, the heart mean dose is reduced from 15 Gy to 7 Gy. The left and right lung V20 are reduced from 27% and 4.4% to 15% and 0.9%. The estimated delivery time is ~ 6 min.
Conclusion: AI with MCTS successfully designed trajectories that are efficient to deliver and dosimetric superior compared with coplanar VMAT.
Contact Email: