EEG is the only tool available that can provide an objective assessment of newborn brain function, in real time, at the cot-side and help guide therapeutic intervention. Over a decade ago the Neonatal Brain Research Group at the INFANT centre began exploring the use of automated analysis of the neonatal EEG. We used machine learning techniques to develop the first robust computer programme for the detection of seizures in newborn babies and this programme completed a clinical investigation in 8 centres across Europe (www.anserstudy.com).
Technology has advanced dramatically in the last few years, and we are now entering a new age of artificial intelligence. There have been exciting developments in the field of machine learning. There is still much we can learn from studying EEG recordings of babies admitted to the NICU and we now wish to use new machine learning techniques to analyse the EEG and study other aspects of neonatal brain health in both preterm and full term neonates. Machine learning requires lots of data for testing, training and validation studies. In this project we will ask parents of neonates undergoing EEG monitoring, if we can keep the EEG data collected for use in our machine learning studies.
Further machine based analysis of EEG’s will expand the knowledge base and help guide further development of computer programmes.
Overall the Neonalysis project aims to explore and use cutting-edge machine learning to identify patterns of brain activity in newborn infants that will help improve diagnosis of neurological disorders, guide individualised treatment strategies, and ultimately lead to better longer term outcomes.