Back to search

Boston University

Modern approaches to sepsis - evolving definitions, clinician roles, and AI-based diagnostic aids

Abstract

dc:description.abstract

Sepsis is an ongoing concern in critical care. It is hard to quickly detect, and rapid deterioration of a patient into septic shock causes death in around 30% - 50% of patients, while survivors may live with organ damage and shorter lifespans. Traditional methods of detection require long laboratory tests and clinician vigilance, which put a strain on hospital resources. New advances in machine learning offer an alternative – using algorithmic analysis in real-time to watch for a deteriorating patient state. The use of readily available data – heart rate, respiratory rate – combined with electronic medical records and fast laboratory tests presents an opportunity for early detection of sepsis, which can potentially make great strides in minimizing damage to patients. A variety of algorithmic methods have been proposed by researchers, and research so far has been promising. Algorithms inretrospective studies have performed equal or better to standard protocols such as SIRS or SOFA. Some promising research even presents the opportunity to approach sepsis diagnosis and treatment in an entirely new manner. At the present stage, however, the field is at too early a stage for use in a clinical environment. This review intends to review some prominent types of machine learning algorithms, as well as discuss current concerns regarding machine learning-based detection support systems (ML-DSS).

Degree

thesis:*
Name thesis:degree_name
Master of Science
Level thesis:degree_level
masters
Discipline thesis:degree_discipline
Medical Sciences
Grantor
Boston University
Year dc:date.issued
2018

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Mathias, Jason
Advisors dc:contributor.advisor
  • Baker, William
  • Flynn, David

Subjects

dc:subject × 7

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivatives 4.0 International
Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/2144/33022

Chain of custody

source
Harvested from
Boston University
Base URL
open.bu.edu/oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Mathias, Jason. Modern approaches to sepsis - evolving definitions, clinician roles, and AI-based diagnostic aids. masters thesis, Boston University, 2018. https://hdl.handle.net/2144/33022