SCH: Wearable Bi-modal Imaging System with Multi-scale AI for Uterine Contraction Mapping in Obstetric Care - Preterm birth, defined as delivery before 37 weeks of gestation, affects ~10% of pregnancies globally and remains the leading cause of neonatal morbidity and mortality. It often leads to long-term complications such as cerebral palsy, developmental delays, and sensory impairments. Despite extensive research, the underlying mechanisms of preterm labor are not well understood, partly due to limitations in current monitoring tools. Clinical devices like tocodynamometry and intrauterine pressure catheters offer limited spatial and temporal resolution and are either invasive or insufficient for early detection of abnormal uterine contractions. This project aims to develop a wearable, AI-integrated, multimodal uterine monitoring system that combines electrical and mechanical signal analysis to offer a more comprehensive understanding of uterine activity during pregnancy and labor. Building on recent advances in electromyometrial imaging (EMMI), the system introduces several key innovations. First, a soft, stretchable, polymer sponge-based sensor capable of capturing both uterine electrical activity and mechanical contractions will be developed, enabling better insight into excitation-contraction coupling in the myometrium. Second, a wireless, motion- artifact-tolerant data acquisition platform will be designed, incorporating embedded machine learning (ML) algorithms for enhanced signal quality and real-time monitoring, even in ambulatory settings. The system will be validated on human subjects against gold-standard clinical tools in labor and delivery settings to assess accuracy, reliability, and usability. Additionally, AI-based predictive analytics will be developed for labor risk stratification and early detection of adverse outcomes, such as preterm birth and fetal distress. By integrating advanced biosensing materials, wireless technology, and AI-driven data interpretation, this project will transform the monitoring of uterine activity, enabling long-term, at-home pregnancy surveillance for high-risk patients. The proposed system has the potential to improve early detection and management of preterm labor, reducing fetal mortality and improving maternal-fetal health outcomes. The long-term impact of this work extends beyond preterm birth prediction; it will advance our fundamental understanding of uterine physiology, facilitate personalized obstetric care, and provide new tools for remote prenatal monitoring. RELEVANCE (See instructions): This project aims to develop a wearable, artificial intelligence (AI)-integrated system for noninvasive monitoring of uterine contractions, combining spatiotemporal mapping of electrical and mechanical signals to improve early detection of labor risks, including preterm birth. By enabling long-term, in-home maternal monitoring, this technology has the potential to enhance prenatal care, reduce fetal mortality, and improve pregnancy outcomes.
