An exploration on methods for early prediction of sepsis

Sepsis is a potentially life-threatening condition that occurs when the body's response to an infection damages its own tissues \cite{Mayo_sepsis_def}. Identification of Sepsis in its early stages is vital in preventing significant organ injury, prolonged hospitalization, and potentially death \cite{mortality_per_hour}. The objective of this thesis is to build a pipeline for early sepsis prediction and examined each steps in the pipeline with the goal to explore different methods and algorithms that can be applied to mitigates the following problems with early sepsis prediction: 1. missingness of data, 2. mismeasurements within data, 3. complex structural relationship between features, 4. imbalance nature of data, and 5. the changing patient states and its corresponding distributions. This thesis had shed lights on the importance of the temporal aspect of medical data on the performance of predictive models in complex medical problems like early sepsis prediction. Further improvements of the prediction pipeline are needed and will be discussed in this thesis.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2022

Author dc:creator

• Chen, Zikun

Contributors dc:contributor

• Sha, Lui R

• sepsis prediction
• health informatics

Statement dc:rights

• Copyright 2022 Zikun Chen

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/115743

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F115743