This repository describes the patient matching process and provides examples of different patient matching approaches.Those approaches and algorithm are implemented in Python using the Python Record Linkage Toolkit. The purpose of this repository is provide an introduction on how to implement patient matching using traditional and modern patient matching techniques. It includes three(3) notebooks introducing and describing the different approaches :
- Introduction to patient matching
- Exploring score-based patient matching
- Patient matching with Unsupervised learning
Each Jupyter Notebook allows you to explore the step by step matching process using synthetic or using your own data immediately see the results if you have a gold standard dataset.
After clicking on the "launch binder" link above, wait for a few minutes to BinderHub build the Docker container.