A recent study conducted by researchers at the University of California San Diego has found that a modified pacifier, paired with AI algorithms, could be used to analyze the mechanics of newborns’ nursing abilities. By measuring variables such as suckling strength and pattern regularity, researchers were able to generate objective data that could potentially prevent the need for surgical interventions in infants experiencing breastfeeding difficulties. Current methods of assessing feeding in infants rely on subjective evaluations by clinicians, but this study aimed to provide a more accurate and objective assessment.
The testing method used in the study involved a device consisting of a simple pacifier connected to a vacuum sensor and a chip for data collection. This device, when connected to a laptop, recorded data as infants sucked on the pacifier. Additionally, machine learning algorithms were utilized to identify abnormalities and outliers in the data, providing clinicians with a more detailed analysis of an infant’s suckling ability. The software developed for this purpose not only displayed data but also compared it with information from other infants to identify patterns and deviations.
Results from the study showed that the device’s data could potentially improve the subjective evaluations currently used by clinicians. By providing rapid and precise data on an infant’s suckling ability early on, clinicians may be able to address underlying issues more quickly and potentially prevent breastfeeding attrition. The study also highlighted the importance of early detection of breastfeeding difficulties, as this critical phase can impact both milk production and breastfeeding success. Ultimately, the goal is to empower clinicians with tools that can improve long-term health outcomes for infants.
In cases where infants are diagnosed with tongue-tie, a condition that limits tongue movement and can affect breastfeeding, clinicians often recommend surgery known as frenotomy. Data from the device in the study suggested that surgical interventions may not be necessary in all cases. For some infants, there was no change in suckling behavior pre and post-surgery. However, for those identified as needing a frenotomy based on abnormal data patterns, the surgery resulted in significantly improved suckling behavior. These findings have important implications for the potential prevention of unnecessary surgical interventions.
The study, approved by UC San Diego’s Internal Review Board, involved healthy full-term infants under 30 days old recruited from various pediatric care centers. Clinicians were blinded to the device data to ensure evaluations were based solely on standard practice. Moving forward, the researchers plan to conduct a clinical trial outside of UC San Diego Health to further validate the device and algorithms. The ultimate goal is to make these tools widely available in pediatric practices, where they can be used during an infant’s first visit to assess breastfeeding abilities more accurately.
The study was funded by various sources, including the Galvanizing Engineering in Medicine initiative at UC San Diego, the National Institutes of Health, and the Willia H. and Mattie Wattis Harris Foundation, among others. By applying statistical analysis and machine learning algorithms to identify infants’ abnormal suckling behavior, the researchers hope to revolutionize the way breastfeeding difficulties are assessed and treated in newborns. The potential to prevent unnecessary surgical interventions and improve long-term health outcomes for infants is a promising result of this innovative study.
