■ Researchers involved

Chang-Min Lee (dlc2048@snu.ac.kr)

Yoonho Na (yoonho94.na@gmail.com)

Kyuri Kim (kyurikim@snu.ac.kr)

■ Description

RPLab is dedicated to pioneering the next generation of Treatment Planning Systems (TPS) for advanced hadron therapy, including carbon-ion and neutron beams. By integrating AI-driven automation with high-performance computing, we aim to transcend the current limitations of radiation oncology, delivering unprecedented precision and personalized care for cancer patients.

1. AI-Driven Automated Segmentation

Precision in radiation therapy begins with an impeccable understanding of patient anatomy. We develop state-of-the-art deep learning architectures to automate the segmentation of tumors and organs-at-risk (OAR). Our technology minimizes inter-observer variability and drastically reduces contouring time, providing a robust foundation for seamless, real-time Adaptive Radiotherapy (ART).

2. GPU-Accelerated Monte Carlo Simulation

To ensure the highest level of dose calculation accuracy, we have engineered a world-first, GPU-accelerated Monte Carlo (MC) simulation engine. While traditional MC simulations are often too slow for clinical use, our engine harnesses massive parallel processing to deliver “gold-standard” accuracy within minutes. This breakthrough enables high-fidelity particle transport simulation, bridging the gap between sophisticated physics and rapid clinical workflows.