Researchers at the University of Helsinki have developed artificial intelligence software which can interpret EEG signals from the brain of a premature infant and evaluate its functional maturity.
One in ten live births is premature and approximately half of the patients are admitted to neonatal intensive care because of preterm birth. The later months of pregnancy witness a rapid brain development in the fetus. However, the health impediments associated with preterm birth can hinder brain development.
Understand the brain development of infants can help design the best possible care and new treatments in preterm births. However, the current methods for such evaluation are neither objective nor precise.
Therefore, Professor Sampsa Vanhatalo, who led the research, utilised the world’s first EEG-based brain maturity evaluation system, the EEG analysis software. It has been developed by Nathan Stevenson, an Australian engineer.
The researchers collected EEG data from 39 preterm infants to feed into the software. The software then calculated hundreds of computational features from each measurement without any intervention from a doctor. These features were combined to estimate the EEG maturational age of the infant. The EEG maturational age estimated by the software was compared with the clinically known true age of the infant. In more than 80% of the cases, the true age of the infant and the computer-generated estimate were within two weeks of one another.
“EEG monitoring combined with automatic analysis provides a practical tool and a first-time opportunity for the monitoring of the neurological development of preterm infants. The information generated will help plan the best possible care for the individual child,” said Vanhatalo.