Electronic Health Record Data in Cardiovascular Research

Abstract

To better understand cardiovascular diseases, patient data from the electronic health record can be used in research. In this thesis, I captured these data manually in clinical registries and developed a research data platform for big data analysis and artificial intelligence. Within this platform, we developed a natural language processing pipeline to automatically classify patients with diagnoses from medical text. Next, we used these data to understand heterogeneity in symptoms, genetic testing and patient outcomes in patients with dilated cardiomyopathy. We developed statistical models and used artificial intelligence to predict life threatening cardiac arrhythmias in dilated cardiomyopathy. In the discussion, I further discuss on how to progress with diagnosis and risk prediction of dilated cardiomyopathy using big data and artificial intelligence.