Simulated annealing applied to IMRT beam angle optimization: A computational study.

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Optimization

Background:

• Beam angle optimization (BAO) is critical for Intensity-Modulated Radiation Therapy (IMRT) planning.
• The non-convex nature of BAO poses significant challenges for traditional optimization methods, often leading to suboptimal solutions.
• Complex head and neck tumors require precise treatment planning to balance target coverage and organ sparing.

Purpose of the Study:

• To evaluate the efficacy of Simulated Annealing (SA) for optimizing beam angles in IMRT for head and neck cancers.
• To determine if SA can improve upon existing methods in achieving both target coverage and organ sparing.
• To assess SA's performance with a fixed, reduced number of irradiation angles.

Main Methods:

• Retrospective application of Simulated Annealing (SA) to ten complex head and neck cancer patient cases.
• Fixed number of beam directions (5 or 7) were used for optimization.
• SA, a probabilistic local search algorithm, was employed to address the multimodal optimization problem of BAO.

Main Results:

• SA demonstrated the ability to find solutions that significantly enhance organ sparing.
• Tumor coverage was maintained even with a reduced number of beam angles.
• The method proved effective in complex clinical scenarios where balancing competing objectives is difficult.

Conclusions:

• Simulated Annealing is a viable and effective approach for beam angle optimization in IMRT for head and neck cancers.
• SA offers a significant advantage in improving organ sparing compared to traditional methods.
• This optimization technique can achieve superior treatment plans with fewer beam angles, simplifying treatment delivery.