Dose prediction of CyberKnife Monte Carlo plan for lung cancer patients based on deep learning: robust learning of variable beam configurations

Abstract

Abstract Background Accurate calculation of lung cancer dose using the Monte Carlo (MC) algorithm in CyberKnife is essential for precise planning. We aim to employ deep learning to directly predict the 3D dose distribution calculated by the MC algorithm, enabling rapid and accurate automatic planning. However, most current methods solely focus on conventional intensity-modulated radiation therapy and assume a consistent beam configuration across all patients. This study seeks to develop a more versatile model incorporating variable beam configurations of CyberKnife and considering the patient's anatomy. Methods This study proposed the AB (anatomy and beam) model to compare with the control Mask (only anatomy) model. These models are based on a 3D U-Net network to investigate the impact of CyberKnife beam encoding information on dose prediction. The study collected 86 lung cancer patients who received the built-in MC algorithm plans of CyberKnife using different beam configurations for training/validation (66 cases) and testing (20 cases). We compared the gamma passing rate, dose difference maps, and relevant dose-volume metrics to evaluate the model's performance. In addition, the Dice similarity coefficients (DSCs) was calculated to assess the spatial correspondence of isodose volumes. Results The AB model demonstrated superior performance compared to the Mask model, particularly in the trajectory dose of the beam. The DSCs of the AB model was 20–40% higher than that of the Mask model in some dose regions. We achieved approximately 99% for the PTV and generally more than 95% for the organs at risk (OARs) referred to the clinical planning dose in the gamma passing rates (3mm/3%). Relative to the Mask model, the AB model exhibited more than 90% improvement in small voxels (P < 0.001). The AB model matched well with the clinical plan's dose-volume histograms (DVHs) and the average dose error for all organs was 1.65 ± 0.69%. Conclusions Our proposed new model signifies a crucial advancement in predicting CyberKnife 3D dose distributions for clinical applications. It enables planners to rapidly and precisely calculate MC doses for lung cancer based on patient-specific beam configurations.