Intern at Innovation Lab for Wearable and Ubiquitous Computing - FAU Erlangen

Conducted in-depth analysis of the MIMIC-IV medical dataset using Python, Pandas, and Scikit-learn to identify key factors affecting ICU patient readmissions, providing insights for improved patient care. Developed a Random Forest model to predict ICU readmissions, achieving 81% accuracy through feature engineering and hyperparameter tuning. Built a secure web application using Next.js and FastAPI to display real-time readmission rates. Integrated key patient features into the web application, improving ICU workflow and decision making by optimizing bed allocation and reducing wait times.