Delay-aware chemotherapy dosing via online critic learning
摘要
This paper proposes a delay-aware adaptive control framework for individualized chemotherapy dosing based on online critic learning. The approach explicitly compensates for pharmacokinetic and pharmacodynamic delays while adapting to patient-specific uncertainties. An online critic network estimates the value function to guide real-time dose adjustments. Simulation results on diverse patient profiles demonstrate effective tumor suppression and toxicity control, highlighting the robustness of the proposed scheme to variations in delay and patient dynamics.