I use data science to lead novel research and bring value through improvements in patient care.What do I do?I engage in the full research cycle from hypothesis inception to implementation. This includes: formulating hypotheses backed by a review of the current literature, experimental design, analytical design, data pipeline construction and validation, analysis using statistical methods and machine learning, the interpretation of findings, authoring manuscripts and presenting findings to clinicians, leaders and other stakeholders. I lead teams of clinicians and assist in all parts of the research cycle. I support resident education by teaching courses on the scientific method, database design, analytical methods and machine learning.Why do I do it?I want to improve healthcare. Healthcare systems have been storing mountains of patient data in the form of electronic medical records, labs, prescription information, imaging in PACs systems, financial data, and various other types. If properly assembled and analyzed, this could be used to gain insight into disease, social determinants of health, medical best practices, population health and epidemiology. This could then be used to create a health system that is agile and adapted to local needs.I want to help physicians engage in the scientific process. Clinicians understand the value of an active research program but often are not trained or have time to be researchers. I step in to help guide them through developing research proposals, regulatory hurdles, experimental design, data access, analysis and manuscript writing. This allows them to realize their full potential as clinicians and keeps them on the cutting edge of medicine. This also benefits our healthcare by helping the organization gain recognition through accreditation with professional organizations like the American College of Surgeons Committee on Trauma (ACS-COT) and the ANCC Magnet Recognition Program for Nursing.What tools do I use?I build data pipelines with SQL, use R and Python for preprocessing and analysis, and bash to schedule jobs. My data sources include data from the electronic health record, claims data, as well as data from the US Census and Tigerline via APIs. My analytical work is primarily carried out using R and Python using RMarkdown documents in RStudio and RStudio Server. These are then compiled as documents and presentations for distribution and manuscript writing.