In recent years, asthma has become increasingly recognised as an umbrella terms for a number of different diseases which present with similar symptoms. We refer to these different diseases as asthma subtypes. However, clinical consensus to define these subtypes is currently lacking.
Many studies have used machine learning techniques in efforts to identify asthma subtypes from data. However, there are a number of limitations in these methods, including (but not limited to) their small sample size. I intend to improve on these methods, and apply them to large electronic health record datasets.
I hope that the results of this study will improve our understanding of asthma heterogeneity by making it possible to assign asthma patients to a subtype based on routinely collected data.
I graduated from the University of Bristol with a BSc inn Mathematics in 2015. Following this I was awarded a Pre-Doctoral Research Methods Fellowship from the NIHR. During this fellowship I obtained an MSc in Medical Statistics from the University of Leicester, and held a year-long post as a Research Associate at the University of Bristol. I started my PhD at the University of Edinburgh’s Centre for Medical Informatics in October 2017.
My research interests are in the application of machine learning methods to medical data. In particular, I am interested in harnessing routinely collected data for research purposes.
This PhD is funded by the Farr Institute and affiliated with the Asthma UK Centre for Applied Research (AUKCAR).