The Opportunities and Challenges for Artificial Intelligence to Improve Sepsis Outcomes in the Paediatric Intensive Care Unit
Abdullah Tarık Aslan, Budi Permana, Patrick N. A. Harris, Kuban D. Naidoo, Michael Pienaar, Adam Irwin
Abstract
Abstract Purpose of Review To describe existing applications of artificial intelligence (AI) in sepsis management and the opportunities and challenges associated with its implementation in the paediatric intensive care unit. Recent Findings Over the last decade, significant advances have occurred in the use of AI techniques, particularly in relation to medical image analysis. Increasingly, these techniques are being applied to a broad array of datasets. The availability of both structured and unstructured data from electronic health records, omics data and digital technologies (for example, portable sensors) is rapidly extending the range of applications for AI. These techniques offer the exciting potential to improve the recognition of sepsis and to help us understand the pathophysiological pathways and therapeutic targets of sepsis. Summary Although AI has great potential to improve sepsis management in children, significant challenges need to be overcome before it can be successfully implemented to change healthcare delivery.