Advancing neonatal health
An interdisciplinary Boston College team was awarded a five-year, $3.2 million grant from the National Institutes of Health to develop the first AI-enabled ultrasound system designed for long-term, point-of-care nutritional monitoring in preterm infants.
The tool would fill a critical gap in assessing the accumulation of lean and fat tissue during infants鈥� neonatal intensive care unit (NICU) stays, when targeted interventions for babies born before 37 completed weeks of pregnancy can have the most significant impact.
The project, 鈥淎dvancing Neonatal Health: AI-enabled Ultrasound for Enhanced Growth and Nutritional Assessment,鈥� will be led by Ferrante Family Assistant Professor of Engineering Bryan Ranger. Co-investigators include Connell School of Nursing Dean Katherine E. Gregory; Associate Professor Jinhee Park, the Deborah Joyce Norberg Faculty Fellow at the Connell School; and Assistant Professor of Computer Science Yuan Yuan.
Clockwise for top left: Ferrante Family Assistant Professor of Engineering Bryan Ranger.; Connell School of Nursing Dean Katherine E. Gregory; CSON Associate Professor and Deborah Joyce Norberg Faculty Fellow Jinhee Park; and Assistant Professor of Computer Science Yuan Yuan. (Ranger and Yuan photos by Caitlin Cunningham; Park photo by Matthew Healey; Gregory photo by Lee Pellegrini)
Additionally, clinical collaborators include Brigham and Women鈥檚 Hospital鈥攐ne of the two founding members of Mass General Brigham, the largest health care provider in Massachusetts鈥攁nd the M Health Fairview Masonic Children鈥檚 Hospital at the University of Minnesota. These sites will be led locally by Katherine Bell, M.D., and Sara Ramel, M.D., respectively.
According to evidence-based clinical database UpToDate, premature babies need significantly higher amounts of calories, proteins, fats, vitamins, and minerals compared to full-term infants to match the rapid growth they would have experienced in the womb. The earlier the preterm infant is born, the fewer stores of fat, protein, iron, and minerals they have, which are essential for respiratory function, heat regulation, and bone development.聽 Enriching a preterm infant鈥檚 nutrition usually addresses the deficit.
鈥淧reterm infants are at high risk for suboptimal growth and neurodevelopmental delays, yet current tools for monitoring nutrition, such as weight, length, and body-mass index (BMI) lack sensitivity to changes in body composition,鈥� Ranger said, noting that BMI is a numerical value calculated from height and weight. 聽鈥淎dvanced methods like air displacement plethysmography鈥攁 test that measures body volume based on air displacement in a sealed chamber鈥攑rovide accurate measures but are costly, impractical for routine use, and infeasible in critically ill infants. Preliminary data from our group demonstrate that ultrasound imaging, when combined with artificial intelligence, can predict body composition with reasonable accuracy in preterm infants.
鈥淭he project鈥檚 innovation is building on this preliminary work to create a scalable, clinically feasible tool that tracks body composition longitudinally, throughout a preterm infant鈥檚 NICU stay and beyond, rather than at a single time point,鈥� he said. 鈥淭he results will improve early detection of growth failure, guide individualized nutritional interventions, and ultimately, reduce the burden of adverse health outcomes in this vulnerable population.鈥�
鈥淲hat is particularly exciting is the potential to bring this kind of information closer to the bedside鈥攖o help clinicians understand not just whether preterm infants are growing, but how they are maturing over time, and ultimately, to support more tailored approaches to nutrition and care.鈥�
Ranger noted that the system will be low-cost, portable, and fully automated, providing a standardized, real-time, non-invasive assessment of lean and fat tissue accrual, a rich amalgamation of features not presently available.
The project鈥檚 primary goals are to validate and standardize ultrasound protocols with real-time feedback mechanisms to ensure high-quality scans; develop and refine AI models that integrate imaging, maternal, and clinical data to predict lean and fat mass with high accuracy; and link early ultrasound-derived nutritional trajectories to neurodevelopmental outcomes to identify at-risk phenotypes and inform targeted interventions.
To achieve these objectives, the team will enroll 160 preterm infants across two sites, collecting ultrasound scans, gold-standard body composition measures during hospitalization, and maternal and clinical data, with follow-up through 24 months after discharge. 聽
Park underscored that weight alone cannot be relied upon as a primary indicator of growth, since it doesn鈥檛 reveal what is happening beneath the surface. 鈥淭he ability to track lean and fat tissue over time could give us a more complete picture of how preterm infants are growing.鈥�
She noted that this interdisciplinary work grew from a 2022 Schiller Institute for Integrated Science and Society-funded pilot study that has evolved into this larger, multi-site project.
鈥淔rom my background in neonatal nursing and research, I鈥檝e seen how challenging it can be to know whether a preterm infant is growing well,鈥� Park said. 聽鈥淲hat is particularly exciting is the potential to bring this kind of information closer to the bedside鈥攖o help clinicians understand not just whether preterm infants are growing, but how they are maturing over time, and ultimately, to support more tailored approaches to nutrition and care.鈥�
Gregory emphasized that preterm infants are small, but the problems they experience are big鈥攖oo extensive for one discipline to solve alone.
鈥淎pproximately half of these newborns experience postnatal growth failure, a major contributor to lengthy hospitalizations and suboptimal health outcomes across the lifespan,鈥� she said. 鈥淭he use of ultrasound and the analytical approach that this team will utilize to address this problem are paradigm-shifting and could lead to novel nutritional strategies that improve preterm infant growth.
鈥淚t鈥檚 been a joy to see this collaboration flourish across campus with the support of multiple schools, departments, and the Schiller Institute.鈥� 聽