FROM RANDOMNESS TO PRECISION: A REVIEW OF MONTE CARLO ALGORITHMS ENHANCING IMRT PLANNING ACCURACY

Abstrak

This review explores the role of Monte Carlo algorithms as a cornerstone of modern external beam radiotherapy, offering unprecedented precision in dose delivery by allowing modulation of beam intensities across multiple angles. This enables optimal tumor coverage while sparing nearby healthy tissues and critical organs-at-risk. However, the accuracy of IMRT is fundamentally constrained by the dose calculation algorithms embedded in the Treatment Planning System (TPS). Conventional analytical methods, such as the Anisotropic Analytical Algorithm (AAA) or convolution–superposition techniques, often rely on approximations that become unreliable in heterogeneous media particularly at tissue-air and bone-soft tissue interfacesresulting in potential dosimetric deviations. To address these limitations, Monte Carlo (MC) algorithms have been increasingly integrated into clinical TPS platforms due to their unparalleled accuracy in simulating photon and electron transport at the particle level. By modeling complex interactions such as scattering and absorption with statistical rigor, MC-based systems provide superior dose calculations, especially in anatomically challenging or high-modulation scenarios. Recent advances in GPU acceleration and AI-assisted simulation have significantly reduced computation times, allowing MC to be deployed in routine workflows without compromising clinical efficiency. Comparative evaluations consistently demonstrate that MC outperforms conventional methods in both accuracy and robustness. In head-and-neck, lung, and prostate IMRT cases, MC has shown superior conformity and organ-sparing capabilities. As a result, MC is transitioning from a validation benchmark to a mainstream clinical tool. With ongoing developments in adaptive radiotherapy and AI integration, MC stands at the forefront of personalized, real-time radiotherapy planning.