Delirium

The Delirium Project

A dashboard for delirium care, enabled by EHR, shows potential to help change trajectory for patients at risk of delirium.

Purpose

Delirium is a widespread and underrecognized medical emergency. It affects up to 80% of ICU patients, 15–25% of hospitalized elderly patients, and 6–30% of non-ICU patients overall. Despite its prevalence, it often goes undetected until it causes serious harm—prolonging hospital stays, increasing mortality, and leading to long-term cognitive decline. Every missed case of delirium is a missed opportunity for prevention and recovery. The Delirium Project uses machine learning and EHR data to predict risk early—offering a scalable, proactive solution to a persistent clinical blind spot.

Features

• Real-time risk flagging using EHR data and AI prediction
• Visual tracking of interventions and care status
• Designed for clinical usability and rapid decision-making

Impact

Enabled proactive, non-drug interventions and improved patient safety.

Screenshots

🚀 Live Demo

Try it now!

Repository

https://github.com/evehliu/Delirium

Table of Contents

• Dashboard
  • 📄 README
  • 📁 PBIX Folder
  • 🖼️ Dashboard Screenshot
• Synthetic Dataset
  • 📚 Dataset Documentation
  • 🐍 Python Script
  • 📊 Sample Data

Skills

SQL   Power BI   ETL/ELT   🧭 Agile   🤝 Stakeholder Engagement