Multi-Modality Assessment of Cardiovascular Outcomes Using Traditional and Machine Learning Approaches

Over the years, stricter and more aggressive health guidelines for reducing cardiovascular disease (CVD) have been put in place. Despite this, CVD remains the leading cause of death in North America. Building risk assessment tools that are more accurate in identifying CVD risk can play a large role in reducing and preventing CVD. These tools include identifying new risk factors, using machine learning approaches to handle large quantities of data and improve predictions of further analyses of data. Furthermore, using traditional statistical methods for optimum interpretability of results, such as identifying individual variables associated with the outcome. The data population studied includes all participants ages 35-79 who have answered questionnaires on lifestyle and health, who provided biospecimens, and additional assessments. The intention is to identify new risk factors and confirm known risk factors for CVD to create prediction models for CVD.