OMOP Medication Deduplication - Project Demo

This video walks through an OMOP-style medication deduplication pipeline that standardizes and deduplicates drug exposure records across multiple healthcare data sources (EHR, Pharmacy, etc.).

The pipeline:
- Maps raw medication text to real RxNorm concept IDs via the Athena vocabulary
- Normalizes dose, unit, frequency, route, and formulation
- Classifies record pairs as duplicate, possible duplicate, or not duplicate using rule-based logic
- Produces a final deduplicated drug exposure table

Built with Python, pandas, rapidfuzz, and a lightweight Streamlit app for visual inspection.

GitHub Repository: https://github.com/sphklt/omop-med-dedup

Tech: Python | OMOP CDM | RxNorm | Athena Vocabulary | Streamlit | rapidfuzz

Видео OMOP Medication Deduplication - Project Demo канала nancy tyagi