SDTDMn0 : a multidimensional distributed data mining framework supporting time series data analysis for critical care research

Premature birth is one of the major perinatal health issues across the world. In 2007, the estimated Canadian preterm birth rate was 8.1 % (CIHI, 2009). Recent research has shown that conditions, such as nosocomial infections or apnoeas, exhibit certain variations in the baby's physiological parameters which can indicate the onset of the event before it can be detected by physicians and nurses. Neonatal Intensive Care Units are some of the highest information producing areas in hospitals. The multidimensional and distributed nature of the data further adds another layer of complexity as physiological changes can occur in one data stream or can be cross-correlated between several streams. With the collection and storage of electronic data becoming a global trend, there is an opportunity to analyse the collected data in order to extract meaningful information and improve healthcare. The aforementioned properties of the data motivate the need for a framework that supports analysis and trend detection in a multidimensional and distributed environment.

Name thesis:degree_name

Master of Health Sciences (MHSc)

Discipline thesis:degree_discipline

Health Informatics

Grantor

University of Ontario Institute of Technology

Year dc:date.issued

2011

Author dc:creator

• Dhanoa, Agam

Advisors dc:contributor.advisor

• McGregor, Carolyn
• James, Andrew
• Calley, Christina

• Distributed data mining
• Temporal abstraction
• Relative alignment
• Time series data analysis
• NICU

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/10155/890

OAI identifier oai:identifier

oai:ontariotechu.scholaris.ca:10155/890