Multiscale Modeling Applications in Cardiovascular Disease and Public Health

Multiscale computational models are powerful tools that integrate data and systems across spatial, temporal, and biological scales in order to make predictions about the behaviors of complex systems. Continued advancements in experimental methods and biomedical technology are generating vast amounts of data that require more sophisticated computational models and analytical methods in order to draw conclusions about complex processes and outcomes that span multiple scales of resolution. In the context of biomedical sciences and human health, these models are particularly relevant in the area of drug design and discovery where small molecules are designed with sub-cellular targets, but have effects across many biological and temporal scales. Experimental data from preclinical animal models can be expansive and have high variability, and multiscale computational models can be leveraged to predict how therapies will translate to humans. Additionally, the rapid growth of big data in health care in recent years including wearable sensors, telehealth, and electronic health records provides an opportunity for multiscale models to integrate disparate data sources to inform evidence-based interventions and transform the delivery of health care. Multiscale computational models provide a unique platform for high-throughput and systematic perturbation of parameters and conditions that may not be otherwise feasible due to time, cost, technological, or ethical considerations. An integrated approach that combines experiments with computational models can aid the design of preclinical and clinical studies, as well as public health interventions. The work presented in this thesis demonstrates novel applications of multiscale modeling approaches to design interventions in the case of diabetic wound healing, infarct healing following myocardial infarction, and the use of electronic health records to identify individual social risk factors and their impact on patient-level health outcomes.