A Bayesian Evaluation of Subtyping Methods in Parkinson’s Disease

Parkinson’s disease (PD) has significant heterogeneity in its presentation. To explain this heterogeneity, several motor subtypes have been proposed. These subtypes make assumptions about how symptoms change over time, the ability to measure symptoms, and the relationships between different symptoms...

Full description

Bibliographic Details
Main Author: Johnson, Andrew Robert
Format: Thesis
Published: Curtin University 2019
Online Access:http://hdl.handle.net/20.500.11937/79166
Description
Summary:Parkinson’s disease (PD) has significant heterogeneity in its presentation. To explain this heterogeneity, several motor subtypes have been proposed. These subtypes make assumptions about how symptoms change over time, the ability to measure symptoms, and the relationships between different symptoms within a given disease subtype. However, current statistical approaches cannot test these assumptions. This thesis used Bayesian statistics to evaluate the assumptions underlying current subtyping methods and developed a new model of PD motor subtypes.