Neotree: Refining a predictive model for the diagnosis of neonatal sepsis in low resource settings
A cross-Faculty collaboration aiming to reduce infant mortality.
![Sepsis foot in hospital. credit DA4554/iStock](/grand-challenges/sites/grand_challenges/files/styles/large_image/public/sepsis.jpg?itok=fPt31yJP)
1 September 2021
Over 9000 babies die daily in low resource settings. 70% of these deaths are preventable through evidence-based guidelines and low-cost interventions. Digital data capture and quality improvement systems delivering clinical management support at the point-of-care provide a potential vehicle with which to (i) encourage adherence to evidence-based guidelines, (ii) provide data and evidence to refine evidence-based guidelines and (ii) improve newborn care and survival. Improved identification and management of newborn sepsis, one of the leading causes of newborn death, could significantly reduce newborn mortality.Â
The team behind this study combines fields of neonatology, clinical epidemiology, biostatistics, and artificial intelligence, and have developed a digital data capture and quality improvement system for newborn care: the Neotree. Through this follow-up project, they aim to explore how machine learning could optimise this existing clinical prediction model.
![](/grand-challenges/sites/grand_challenges/files/styles/large_image/public/p1040650.jpg?itok=a9VknMH1)
The NEOtree project exhibition panel at the Grand Challenges, Grand Impacts exhibition, 2023