Speaker
Description
Achieving an optimal balance between effective tumor control and healthy tissue protection remains a key challenge in radiotherapy. Conventional treatment planning methods are often limited to simplified mathematical models and cannot fully account for patients’ anatomical complexities and dynamic tissue responses.
In this study, we present an innovative approach integrating Monte Carlo simulations with machine learning algorithms to predict radiation dose distributions in tumors and surrounding healthy tissues with high accuracy. Clinical datasets, including CT/MRI images and in vivo dose measurements from 30 anonymized cases, were used to train personalized models. The machine learning algorithm refines Monte Carlo simulation results by analyzing complex dose patterns, resulting in more precise dose predictions. Monte Carlo simulations served as the reference standard for dosimetric validation.
Preliminary results demonstrate that this approach reduces dose estimation errors by up to 20%, significantly decreases radiation exposure to healthy tissues, and enhances safety for both patients and clinical staff. This AI-enhanced methodology enables precise radiotherapy planning for anatomically complex cases and offers potential applications across broader areas of medical dosimetry.
Keywords:
Dosimetry, Radiotherapy, Medical Physics, Monte Carlo Simulation, Artificial Intelligence, Radiation Protection, Personalized Treatment.