Your address will show here +12 34 56 78

Medical Imaging

January 2024 - January 2027

Chimeric Antigen Receptor (CAR) T cell therapy has emerged as a promising treatment modality for various forms of cancer, harnessing the patient's own immune system to target and destroy cancer cells. This project focuses on the control of the fabrication process of CAR T cells through the application of reinforcement learning (RL) and federated learning techniques. The current manufacturing processes for CAR T cells are often labor-intensive, time-consuming, and subject to variability, leading to challenges in scalability, consistency, and cost-effectiveness. Traditional approaches rely heavily on manual intervention and fixed protocols, resulting in inefficiencies and suboptimal product quality. In pursuit of scalability, a multi-site production approach as been choose in this project. We explore the solutions proposed by Federated learning to tailor optimization efforts to each site's unique data while ensuring quality standards. To tackle the challenges of consistency and cost-effectiveness, we've turned to reinforcement learning to refine the fabrication process of CAR T cells. RL is used to train an agent in formulating optimal decision-making strategies through dynamic interaction with its environment, thereby fostering continuous improvements in manufacturing efficiency and product quality.