Newsroom

Using big data to help the smallest babies

Canadian researchers apply machine learning to newborn blood tests to help identify preterm infants in low-resource countries

janvier 27, 2023

Dr. Steven Hawken“Every newborn blood sample contains a metabolic fingerprint and by developing machine learning models using these fingerprints, we can estimate gestational age." - Dr. Steven HawkenPreterm birth is a leading cause of death among children under the age of five, with low resource countries facing the greatest challenge.

There are ideas and programs to address this challenge, but to know if they are working, researchers need to be able to accurately measure changes in preterm birth rates at a population level. This has been particularly difficult in countries where prenatal ultrasound is not routinely available for estimating gestational age.

New research led by Drs. Steven Hawken and Kumanan Wilson shows that a mathematical model that reads the “metabolic fingerprint” of newborn blood samples could help. After initial research in Canada, Bangladesh and other countries, the team recently completed an external validation study in Kenya involving more than 1,000 newborns.

As reported in PLOS Global Public Health, the mathematical model was accurate in identifying gestational age and preterm births and the blood testing method was feasible.

Dr. Kumanan Wilson “Routine newborn blood screening can also reveal treatable genetic or metabolic problems,”-Dr. Kumanan Wilson “Every newborn blood sample contains a metabolic fingerprint and by developing machine learning models using these fingerprints, we can estimate gestational age,” said Dr. Hawken, senior scientist at The Ottawa Hospital and associate professor at the University of Ottawa. “This kind of approach could be crucial to global efforts to reduce preterm birth and improve newborn health.”

“Routine newborn blood screening can also reveal treatable genetic or metabolic problems,” said Dr. Wilson, senior scientist and internal medicine specialist at The Ottawa Hospital and professor at the University of Ottawa. “We are continuing to work with partners in several countries, and with Ontario’s Newborn Screening Lab at CHEO, to find ways to improve access to this technology.”

Authors: Steven Hawken, Victoria Ward, A. Brianne Bota, Monica Lamoureux, Robin Ducharme, Lindsay A. Wilson, Nancy Otieno, Stephen Munga, Bryan O. Nyawanda, Raphael Atito, David K. Stevenson, Pranesh Chakraborty, Gary L. Darmstadt, Kumanan Wilson.

Funding: Bill & Melinda Gates Foundation. All research at The Ottawa Hospital is also enabled by generous donors to The Ottawa Hospital Foundation.

Core Resources and Partners: Ottawa Methods Centre, CHEO, Newborn Screening Ontario, Kenya Medical Research Institute, Stanford University

The Ottawa Hospital is a leading academic health, research and learning hospital proudly affiliated with the University of Ottawa and supported by The Ottawa Hospital Foundation.

 

Disease and research area tags: Newborn and child health, Big data