This project provides a fully reproducible real-world evidence study examining 28-day mortality outcomes in ICU sepsis patients treated with vasopressors versus fluid resuscitation alone. It demonstrates an end-to-end clinical data science workflow, from raw EHR data ingestion to survival analysis and SHAP interpretability. The project is designed for clinical researchers and data scientists working with electronic health record data. It offers a thoughtful approach to analyzing treatment outcomes and provides a polished workflow for similar studies.
View on GitHub →leen01/mimic_iv