Prenatal care is an essential preventive service aimed at improving health outcomes for pregnant individuals. The American College of Obstetricians and Gynecologists (ACOG) has recently introduced guidelines advocating for tailored prenatal care, known as PATH (Plan for Tailored Healthcare). We present the PATHFinder Agent (Planner for Appropriate Tailored Healthcare), an end-to-end conversational agent system that collects patient health and social context through structured dialogue, curates individualized prenatal care plans aligned with PATH guidelines, and surfaces community resources from Michigan 211.
The system features a four-stage workflow that spans patient intake, dynamic interaction, plan synthesis, and clinician oversight. We evaluate frontier large language models (LLMs) across five clinical dimensions and find that GPT-5.2 achieves the highest average score (77.6%) while identifying key gaps in antenatal testing recommendations. We discuss future validation through human participant studies and randomized controlled trials.
Blogger's Review: This system showcases the potential of AI in healthcare, particularly in personalized medicine through conversational interactions. However, ensuring data privacy and security in real-world applications remains a critical challenge. Future research should focus on the interpretability and clinical applicability of the model.