Metaheuristic-Based IMRT Optimization for Lung Cancer Treatment under Respiratory Motion

Abstract

Introduction: Intensity-Modulated Radiation Therapy (IMRT) is a widely adopted cancer treatment technique due to its precision in delivering radiation doses to tumors while minimizing exposure to surrounding healthy tissues. A major challenge in IMRT, however, is the movement of tumors and organs during treatment, primarily due to respiration. This motion can lead to inadequate tumor dosing or increased radiation to nearby healthy organs.

Objectives: This research aims to improve IMRT planning by incorporating motion-aware optimization techniques that account for respiratory-induced tumor and organ motion. Specifically, the study aims to identify the most effective metaheuristic optimization algorithm for enhancing IMRT performance and treatment safety.

Methods: Three metaheuristic algorithms—Genetic Bee Colony (GBC), Starling Murmuration Optimization (SMO), and Walrus Optimization Algorithm (WaOA)—were employed to optimize the IMRT objective function, which was modified to consider respiratory motion. The performance of IMRT without optimization was compared with IMRT optimized using each of these algorithms. The Dose Volume Histogram (DVH) was used as the evaluation metric to assess radiation dose distribution across the tumor and surrounding organs.

Results: Among the tested methods, the SMO-based IMRT delivered the prescribed dose of 72.7 Gy to the tumor while significantly reducing radiation exposure to critical organs such as the lung, spinal cord, and heart. Unlike unoptimized IMRT and the other two optimization techniques (GBC and WaOA), SMO demonstrated superior performance in maintaining tumor dose coverage and minimizing collateral damage.

Conclusions: The integration of the Starling Murmuration Optimization algorithm into IMRT planning offers a more effective and safer approach for treating tumors affected by respiratory motion. SMO-based IMRT enhances dose precision and reduces radiation to healthy tissues, making it a promising strategy for lung cancer therapy under motion uncertainty.