Infant Nutrition and Machine-based Learning: Preliminary findings reveal promise

“This electronic brain is as good as a parent’s ear!”

“The data is incredibly exciting! It is like moving from the X-ray to MRI in terms of its significance”,

the remarks given by Thomas Ludwig, Principal Scientist Paediatric Gastroenterology at Danone Nutricia Research; when he was interviewed by NutraIngredients-Asia about creating the next generation of infant nutrition in Precision Nutrition D-Lab.

Inconsolable crying and fussing are hallmarks of infant colic and are of relevant concern to parents and pediatricians. There was until now no reliable method to automatically detect and quantify crying and fussing for diagnostic or for clinical study purposes, for example to measure the effectiveness of interventions.

Machine-based learning can help to close this gap.

In collaboration with LENA Foundation, Danone Nutricia Research utilised the LENA (Language ENvironment Analysis) system to automatically identify, quantify, and distinguish periods of crying versus fussing in a pilot sample of infants.

The preliminary findings from a pilot study in infants were promising. The system is able to distinguish between crying and fussing with 90% sensitivity, 92% specificity, and 91% overall accuracy.

Machine-based learning could be as effective as the human ear when it comes to detecting infants crying or fussing. This has opened up possibilities for potential clinical applications such as identifying and characterising infantile colic.

Read more about the interview of NutraIngredients with Thomas Ludwig, Principal Scientist Paediatric Gastroenterology at Danone Nutricia Research.


About NutraIngredients
NutraIngredients Asia is a website that provides the latest news to decision makers in the areas of supplements, health and nutrition across the Asia-Pacific region.