Early Detection of Sepsis in Neonatal and Adult Patients Using Machine Learning Approaches - Aiding Diagnosis for Clinicians

This thesis systematically reviewed the relevant literature, summarised the current research status in computer-aided sepsis detection, and proposed a unified sepsis detection framework. Within this framework, we explored multiple methods, including classic machine learning and deep learning models, in the task of early detection of sepsis in infants and adults, and achieved promising outcomes that are able to assist clinicians in their decision making during the diagnostic process.