Research
Columbia University to Lead ARPA-H-Funded Effort to Transform Childbirth Through AI-Powered Maternal–Fetal Monitoring
The interdisciplinary group joins a national effort to transform obstetric care with intelligent sensing and real-time clinical decision support.
A Columbia-led team has received a Phase I award of up to $5 million from the Advanced Research Projects Agency for Health (ARPA-H) to develop HARMONI—an AI-powered wearable monitoring system designed to give clinicians continuous, objective insight into the health of mother and baby during labor. Columbia will serve as the lead coordinating institution. The project is one of eight selected in the inaugural year of ARPA-H's Making Obstetrics Care Smart (MOCS) program, a four-year initiative to make the United States the safest place in the world to give birth.
“This project has a very special place in my heart,” said Xia Zhou, associate professor of computer science at Columbia Engineering, who serves as co-lead principal investigator and technical co-lead. “As a mother of two, I have experienced firsthand how stressful and uncertain labor can be, even with excellent medical care.”
Zhou’s longtime collaborator, Tam Vu, Thomas A. and Georgina Tugwell Russo 1977 Distinguished Professor in Computer Science at Dartmouth College, will serve as lead principal investigator.
A new vision for obstetric care
The technology used to monitor babies during labor has changed remarkably little since electronic fetal monitoring was introduced in the late 1960s. Today, clinicians caring for women in labor still rely primarily on fetal heart rate tracings that can be difficult to interpret consistently. Studies have shown that different clinicians reviewing the same tracing may reach different conclusions, creating uncertainty in one of the most consequential moments in medicine. Despite the United States spending more per capita on maternal care than any other high-income country, maternal and neonatal outcomes remain among the poorest.
The new effort aims to change that.
The HARMONI platform combines a wireless abdominal smart belt with a companion ear-worn sensor. Together, they continuously monitor fetal cardiovascular activity, uterine contractions, fetal movement, maternal electrocardiogram, oxygen saturation, respiration, temperature, and other physiological signals. Rather than evaluating each signal independently, AI algorithms analyze the dynamic interactions between mother and fetus, providing clinicians with objective assessments of fetal well-being, insights into likely causes of fetal compromise, and information to support timely intervention.
“Current fetal monitoring provides only part of the picture,” says Matthew Hoffman, vice chair of research in the Department of Obstetrics & Gynecology at Vagelos College of Physicians and Surgeons and an expert in labor management and fetal monitoring. “By combining maternal and fetal physiology with AI-driven analysis, HARMONI has the potential to provide clinicians with more actionable information when timely decisions matter most.”
The award's first phase focuses on the project's technical backbone: the belt and ear sensor and the software that fuses their signals. This phase will also include an early conversation with the U.S. Food and Drug Administration (FDA) about a path to approval, along with a small feasibility study. Later phases could carry the system toward an FDA submission.
Combining strengths
Columbia University serves as the lead coordinating institution for HARMONI and brings together investigators from the University’s Departments of Computer Science, Obstetrics & Gynecology, and Psychiatry. In addition to Zhou and Hoffman, the Columbia team also includes developmental psychobiologist William Fifer and biomedical engineer and developmental neuroscientist Nicolo Pini, whose expertise spans fetal physiology, neurodevelopment, and physiological signal processing.
The project exemplifies collaboration across engineering, medicine, and behavioral science. Zhou's laboratory has pioneered unobtrusive wearable sensing technologies for maternal and infant health. Its previous award-winning Joey project developed a fabric-based necklace that supports Kangaroo Mother Care while continuously monitoring newborn vital signs without attaching sensors directly to the infant. HARMONI extends this broader vision by bringing intelligent sensing and AI from neonatal care to labor and delivery. Hoffman provides clinical leadership and ensures the technologies address the realities of labor and delivery. Fifer and Pini contribute expertise in joint analysis of maternal-fetal physiological signals, fetal and neonatal neurodevelopment, developmental neuroscience, and physiological signal processing.
At Columbia, researchers will lead the development of the maternal abdominal sensing belt, multimodal AI algorithms, maternal-fetal concordance analysis, and clinical coordination. The interdisciplinary collaboration reflects Columbia's strength in translating advances in engineering into technologies that directly improve patient care.
Looking ahead
The Columbia team will collaborate closely with Dartmouth College, the University of Massachusetts Amherst, clinical partners, and industry collaborators throughout the project. Tam Vu at Dartmouth College leads the design of the system backbone architecture. Phuc Nguyen at the University of Massachusetts Amherst leads the engineering that links the wearable sensors to hospital systems. John Yeh, professor emeritus at Harvard Medical School, advises on clinical strategy. The team plans to engage with three types of clinical sites: Columbia University Irving Medical Center (urban), Geisinger (rural), and the Gentle Landing Birth Center.
For Columbia, the project represents an opportunity to demonstrate how close collaboration across engineering, obstetrics, psychiatry, and artificial intelligence can address one of healthcare's most challenging problems. The researchers envision a future in which labor decisions are guided not by uncertainty, but by continuous, objective evidence that improves outcomes for mothers and babies alike.
About HARMONI
HARMONI (Hypoxia Assessment via Real-time Maternal-fetal Oxygenation and Neurophysiological Integration) is supported in part by ARPA-H through the Making Obstetrics Care Smart (MOCS) program. The consortium includes Columbia University, Dartmouth College, the University of Massachusetts Amherst, clinical partners, and industry partners working together to develop next-generation maternal-fetal monitoring technologies.
This research was funded, in part, by the Advanced Research Projects Agency for Health (ARPA-H). The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S. Government.