Current students

Electronic health records (EHRs) have dramatically transformed how healthcare is delivered and managed, providing a wealth of data that can be used to improve patient outcomes and population health. This project will explore the potential of EHRs for monitoring and modelling health-related biomarkers and behavioural trajectories using functional data analysis (FDA).
Functional data analysis is particularly useful for analyzing longitudinal data, such as EHRs, where measurements are taken over time and are interdependent and unregular. The proposed models will evaluate the possible association between trajectories (or groups of trajectories) with various diseases, focusing on cardiac events and mental decline.
One of the great sources of EHRs is the ability to track medication adherence, which is a critical component of disease management. By analyzing drug adherence trajectories, the models developed in this project can identify patients at risk of poor health outcomes due to non-adherence to medication regimens. This information can then be used to develop targeted interventions to improve adherence and patient outcomes.
In addition to drug adherence, the proposed models will examine biomarker trajectories, such as blood pressure and cholesterol levels. By analyzing these trajectories, the models can identify patterns that may be predictive of future health outcomes, allowing for early interventions.
The models developed will also explore the determinants that affect the progress of trajectories. These determinants may include demographic and socioeconomic factors, such as age and gender, as well as clinical factors, such as comorbidities. By identifying the most important determinants that affect the progress of trajectories, healthcare providers can develop personalized treatment plans tailored to individual patients.
Overall, the use of EHRs for monitoring and modelling health-related biomarker and behavioural trajectories has excellent potential for improving patient outcomes and population health. By developing accurate models to identify patients at risk of poor health outcomes, healthcare providers can implement targeted interventions that improve adherence, reduce the risk of cardiac events, and improve patient health.