Abstract
Heart Failure (HF) is a highly prevalent and deadly chronic condition, with high mortality rates and expensive treatments around the world. HF healthcare can be improved through automated Machine Learning (ML) systems, to assist medical experts in diagnosis and decisions. ML systems depend on high quality data. Class imbalance and missing data are major problems of real-world data. My thesis addresses these data issues, with a focus on assisting HF patients of Western Australia. I have developed improved ML models to predict 30-day HF readmissions, deal with class imbalance, and effectively impute missing values in data.
Original language | English |
---|---|
Qualification | Doctor of Philosophy |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 14 May 2021 |
DOIs | |
Publication status | Unpublished - 2021 |