This paper introduces an innovative approach to modeling and optimizing biomedical systems, with a focus on the superovulation stage of in vitro fertilization (IVF) and the brain disease encephalitis. Natural particulate processes, such as crystallization, involve particle growth models where particle size distribution evolves over time. These processes can be effectively regulated using optimal control theory. Drawing an analogy between such particulate processes and superovulation in IVF, we developed a dynamic model for follicle size distribution (FSD) in IVF. Using this model, we applied optimal control theory to derive patient-specific optimal drug dosage profiles. Results from four clinical trials and a year-long cohort study involving 950 patients demonstrated that utilizing the Opt-IVF platform for hormone dosing significantly reduces hormone usage, costs, and side effects. Importantly, it eliminates the need for ultrasound testing after day five, increases embryo yields, and boosts pregnancy rates by over 40% on average compared to controls. These benefits were particularly notable in older patients and poor responders, underscoring Opt-IVF’s potential to transform clinical practices.
Building on this success, we are collaborating with physicians at the University of Pittsburgh Medical Center to address encephalitis, a severe central nervous system (CNS) disease caused by Herpes simplex virus 1 (HSV-1). HSV-1 has a strong affinity for neurogenic niches in the CNS, especially those rich in neural precursor cells (NPCs), which are crucial for neurogenesis. Using a similar modeling approach to Opt-IVF, we developed an NPC growth model that integrates the effects of HSV-1 and therapeutic interventions. This research holds promise for advancing personalized treatment strategies for encephalitis patients, paving the way for improved clinical outcomes.
