TU-G-BRB-01: Continuous Path Optimization for Non-Coplanar Variant SAD IMRT Delivery Using C-Arm Machines

Area of Science:

• Medical Physics
• Radiation Oncology
• Robotics in Medicine

Background:

• Non-coplanar Intensity-Modulated Radiation Therapy (IMRT) offers potential dose distribution benefits.
• Current methods struggle with optimizing complex paths for IMRT beams with varying Source-Axis Distances (SADs).
• Ensuring collision-free and efficient machine movement is critical for advanced radiotherapy systems.

Purpose of the Study:

• To develop and investigate a continuous path optimization methodology for traversing non-coplanar IMRT beams with variant SADs.
• To orchestrate couch and gantry movements for zero-collision, minimal patient motion, and reduced machine travel time.
• To enable safe, accurate, and efficient delivery of advanced IMRT techniques.

Main Methods:

• A tumor-centric coordinate system was used to represent the traversing path as a continuous curve.
• Optimization involved a composite objective function penalizing patient motion and travel time while ensuring beam coverage.
• Level set representation and variational flows were employed for optimization within a defined feasibility manifold (collision-free space).

Main Results:

• The proposed method generated smooth and intuitively appealing delivery paths.
• Optimized paths outperformed generic Traveling Salesman Problem (TSP) solutions under relaxed conditions.
• Results aligned with specially tuned TSP solutions, validating the approach.

Conclusions:

• A novel systematic approach was developed for automatically determining continuous paths for non-coplanar, varying SAD IMRT beams.
• The method accommodates patient-specific collision zones and ensures continuous avoidance.
• This facilitates safe, accurate, and efficient non-coplanar IMRT delivery using advanced robotic C-arm systems.