Files in this item



application/pdfSP20-ECE499-Thesis-Chen, Zikun.pdf (295kB)Restricted to U of Illinois
(no description provided)PDF


Title:Pathophysiological model for early detection of sepsis
Author(s):Chen, Zikun
Contributor(s):Lui, Sha
Subject(s):medical guidance system
preventable medical error
pediatric sepsis
pathophysiology model
data-driven early warning system
Abstract:Pathophysiology Centric Early Detection and Treatment Guidance System is a cyber medical early warning system which assists the early detection of the onset of serious adverse events by guiding medical staff to take preemptive interventions. This guidance system is a pathophysiologic model-based and data-driven early warning system: it keeps track of multiple pathophysiological processes, their interactions, and updates the patient state based on the requested information obtained in real time. The objective of this thesis is to examine a Bayesian method to achieve a reliable prediction by transforming a Boolean decision tree that is commonly used for diagnosis to a trustable and interpretable probability based graph that can be used for early detection of pediatric sepsis and visualization for medical practitioners. This thesis compares the current diagnosis methods of sepsis, shares the results of Bayesian implementation, and proposes possible avenues for future early detection tools that can be used in a medical guidance system. The approach used in this thesis is to transform the decision tree used for Systemic Inflammatory Response Syndrome (SIRS) into a simple Bayesian model. Further improvements of the model are needed and will be discussed in this thesis.
Issue Date:2020-05
Date Available in IDEALS:2020-06-12

This item appears in the following Collection(s)

Item Statistics