Paper Contents
Abstract
Chimeric Antigen Receptor T-cell (CAR-T) therapy has revolutionized personalized cancer treatment, offering significant benefits for patients with hematological malignancies. However, the complex supply chain of CAR-T therapy is highly susceptible to inefficiencies in slot allocation, leading to delays and treatment cancellations. Machine learning (ML)-driven predictive modeling has emerged as a promising approach to optimize scheduling, enhance decision-making, and reduce cancellation losses. This paper evaluates existing ML applications in healthcare logistics, identifies key challenges in CAR-T therapy scheduling, and proposes a Predictive CAR-T Slot Optimization Framework (PCS-OF) integrating supervised learning, reinforcement learning, and explainable AI techniques. Experimental simulations demonstrate that reinforcement learning significantly outperforms traditional scheduling methods in reducing cancellations and improving slot utilization. Despite its potential, challenges such as data quality, model interpretability, and computational efficiency must be addressed for real-world deployment. Future research should focus on standardized data frameworks, explainable AI models, and fairness-aware ML algorithms to ensure equitable and effective integration of predictive analytics into CAR-T supply chains.
Copyright
Copyright © 2025 Nidhi Shashikumar. This is an open access article distributed under the Creative Commons Attribution License.
