Major project Healthcare & Hospitals

Readmission risk model in production

Build a readmission model and deploy it with de-identified data, audit logs and monitoring suitable for a hospital.

The brief

A hospital wants readmission risk on every discharge, but its IT team needs privacy, audit trails and proof the model still works next year. Build the model and deploy it with de-identification, audit logs, versioned models and monitoring. Good looks like: a reproducible pipeline, a deployed API with audit logs, a release gate on model quality, and a drift and fairness report.

What you will build

Build and evaluate the readmission model
De-identify data and document what is stored
Register models and serve them from an API in Docker
Log every prediction with model version for audit
Monitor drift and performance across patient groups
Write a model card for the hospital

Free, open data to build it with

Diabetes 130-US hospitals for years 1999–2008 ↗ UCI Machine Learning Repository

Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.