University of Nottingham

Machine Learning to support care for premature babies in Low-Income Countries

2017 Newspaed-gestation-age-2

Researchers from The University of Nottingham have won the Best Conference Paper at the prestigious 12th IEEE Conference on Automatic Face and Gesture Recognition (FG2017) in Washington DC.  Submitted papers were reviewed by over 250 experts to decide on the winner. 

The paper, presented orally by Dr Mercedes Torres-Torres, described the clinical trial, GestATion, undertaken in Nottingham aiming to use machine learning to estimate gestation age in newborn babies using still images of the babies foot, face and ear.

Show more

The method could be used in low-middle income countries where antenatal care is poor and many babies are born prematurely with no record of this or unable to decide on the care they require.  The team aim to develop a software App that uses the smartphone camera and internet connectivity to support the healthcare worker looking after the baby with gestation specific advice and provide the region with detail mapping of all births. 

Dr Don Sharkey, Associate Professor of Neonatal Medicine and chief investigator, Dr Caz Henry, Carole Ward (all from Academic Child Health) and Dr Michel Valstar (Lecturer in Computer Science) were also authors of the paper.  The study was funded by the Bill and Melinda Gates Foundation.”


Find out more about the GestATion project