The American Recovery and Reinvestment Act mandated that all healthcare providers incorporate and utilize electronic medical/health records (EMR/EHR) by 2014. Due to this regulation, more than 96% of healthcare providers in the United States have adopted EHRs. These datasets contain a wealth of information on patient diseases, diagnoses, treatment decisions, and outcomes. By mining this information, we could make more accurate diagnoses earlier, make better treatment decisions, provide better care to marginalized patient populations that are understudied in trials, and make meaningful progress towards achieving healthcare equity.
Scope: Our project is focused on inflammatory bowel disease (IBD) and we aim to apply natural language processing (NLP) to extract key patient information, and study differential patient responses to a variety of FDA-approved medications.