Cardiovascular Disease Risk Prediction

Libraries: PySpark , Pandas , Scikit-learn , Keras (TensorFlow) , Plotly , XGBoost

In my master thesis I researched the benefits of applying deep learning techniques to cardiovascular disease risk prediction. A literature review was conducted first to identify and overview current approaches that are used for cardiovascular disease risk prediction in practice. Then, alternative machine learning techniques were selected that could lead to an increase in predictive performance compared to earlier reported techniques. Also, robustness of the generated prediction models were assessed. The data was provided by the UK Biobank.

The thesis was supervised by Georg Krempl and Ayoub Bagheri.