Makayla Roma

Masters student at The University of Michigan majoring in Industrial and Operations engineering with a focus on data analytics. 

Center For Healthcare Engineering and Patient Safety, University of Michigan May 2021 - Current Data Scientist
- Improved access to lung cancer screening with 435 additional patients being screened each year through the devel- opment of a workforce requirement dashboard which streamlined screening workflows and identified optimal staffing needs, using VBA and Tableau

-Managed a team of six student researchers, collaborating with clinical stakeholders of the lung cancer screening pro- gram, to provide organizational visibility at the VA Ann Arbor
- Accessed patient records from the VA’s Patient Health API in Python, analyzing these records with SQL queries to determine the average time to screen different patient groups

Veterans Health Administration, Ann Arbor May 2022 - August 2022 Healthcare Analytics Intern
- Increased capacity of Ann Arbor’s VA Hepatology Clinic by 12% and decreased patient mortality by 16%, by developing a patient scheduling risk triaging system using linear programming

- Coded a discrete event simulation model in Python to compare the use of a risk triaging system to current clinical scheduling, which demonstrated that the use of a risk triaging system allows 12 more patients to be scheduled each week

Regenstrief Center for Healthcare Engineering, Purdue University May 2020 - August 2020 Research Fellow
- Developed a logistic regression model, with an accuracy of 81% in R, to determine the impact of state home healthcare policies on Medicaid spending

- Identified that states who implement control policies are two times more likely to have spending increases and three times more likely to have poorer health outcomes
- Presented results to Indiana CMS representatives to inform state Medicaid policy decisions

ISEED, University at Buffalo May 2019- August 2019 Data Science Intern
- Implemented and compared the performance of Random Forest, Logistic Regression, CART, Boosting, and Bagging models to predict who is most likely to develop heart disease in R with an accuracy of 86%

- Utilized data from three different sources (CDC BRFSS, Dartmouth Atlas, and USDA Food Environment Atlas), linking these data sets by zip code in R

University of Michigan August 2021 - April 2023 Master of Industrial and Operations Engineering, Graduate Certificate in Data Science

• GPA: 3.9/4.0, National Science Foundation Graduate Research Fellow, Rackham Merit Fellow, OSTEM

SUNY at Buffalo August 2016 - May 2021 B.S. Industrial Engineering , B.A. Economics, B.A. Political Science

• GPA: 3.9/4.0, Daniel Ackers Scholar, Harold O. Wolf Award Recipient, Tau Beta Pi Member