Using AI to Revolutionize Feeding Practices in the NICU—with Spark Fund Awardee Professor Emily Zimmerman

Jan 18, 2023 | Innovation, Spark Fund

Many processes come automatically to most adults—such as eating and drinking. However, to an infant, feeding can be quite a complicated task. Infant feeding requires the careful coordination of sucking, swallowing, and breathing. This coordination can be a considerable challenge to newborn babies learning these skills in real-time, especially those born preterm.

Around 10% of all live births are preterm, and nearly all preterm infants struggle to feed. Approximately 2.8 million infants face feeding challenges in the US each year, and these numbers are rising.

Current infant feeding assessments tend to be a subjective, trial-and-error process. It is up to the bedside nurse to decide when an infant is ready to feed. If the clinician is incorrect about the infant’s readiness, the infant can choke or aspirate fluid into their lungs. This can lead to feeding aversions, prolonged hospitalizations, and a slew of long-term negative effects on the child, including poor growth, issues gaining weight, developmental delays, and the overall inability to thrive.

Therefore, there is a massive need for a universal and affordable system to examine sucking and feeding in preterm infants quantitatively. That’s why Professor Emily Zimmerman and her team at NeuroSense Diagnostics—a Northeastern spin-out company—created the only non-contact infant suck monitoring system.

This work has earned them selection as one of the Spring 2022 Spark Fund awardees.

The First and Only Non-Contact Infant Suck Monitoring Technology

NeuroSense Diagnostics is a smartphone app that revolutionizes the field of infant suck monitoring. It is the first and only contactless monitoring system, so it does not require the use of any bulky, expensive equipment. The non-contact technology enables a more natural analysis because the infant can be in a comfortable environment, such as in their crib or snuggling with mom. It is also more affordable and convenient because clinicians can use it anywhere at any time.

NeuroSense Diagnostics analyzes infant suck using cutting-edge computer vision technology, artificial intelligence, and machine learning to conduct facial gesture and landmark tracking analysis. Before every feeding session, a clinician uses the app to take a 2-minute video of the infant while using their preferred pacifier. The app allows all clinical staff to upload a video of the infant sucking, instead of limiting data collection to only the clinicians trained on a certain device. The app also allows clinicians to monitor patients remotely, which no other existing technology can do.

This method of objectively measuring and longitudinally monitoring infant suck patterns can be considered the first line of attack in understanding the suck-swallow-breathe mechanism in infants.

“The more we are learning about infant suck, the more our data show that it relates to many important outcomes for a child, such as growth, feeding, and neurodevelopment,” says Zimmerman.

Commercialization with the Spark Fund

Through commercialization, the ultimate goal of NeuroSense Diagnostics is that everyone has access to the technology. In addition, the team hopes that the simple two-minute infant video assessment will serve as an early window into the infant’s developing central nervous system.

“My ideal future outcome is to have our NeuroSense Diagnostic Application be part of everyone’s newborn care prior to hospital discharge, much like an initial hearing screen has become part of infant assessment,” says Zimmerman. “This can help to guide clinical decisions and allow families to access resources earlier.”

With the help of the Spark Fund, the team will be able to create their prototype mobile phone application.

“In research, you are the expert within your domain, but in commercialization, you must hear from the user and adapt accordingly,” says Zimmerman. “Commercialization is a journey, and the skills are not learned overnight. This was my second time applying for the Spark Fund. Going through the process helped shape the product, enabled me to think of direct impact to the users and the infants, and translated my research ideas into a commercial product.”

As a first-of-its-kind smartphone app, Neurosense Diagnostics hopes to shift how clinicians approach feeding practices in the NICU and beyond.

Learn more about Professor Zimmerman’s research and the other Spark Fund Award grantees here.

Written by Elizabeth Creason