Course Description
Advanced analytics techniques for healthcare operations including predictive modeling, prescriptive analytics, and machine learning applications. Students work with real healthcare datasets to develop solutions for clinical and operational problems. Covers data governance, model deployment, and ethical considerations.
Credit Hours
3 credits (3 lecture with lab)
Prerequisites
Statistics course, programming experience (Python or R)
Student Learning Outcomes
- Develop predictive models for clinical outcomes (readmission, mortality, LOS)
- Apply machine learning algorithms to healthcare classification problems
- Design prescriptive analytics solutions for resource allocation
- Handle healthcare data: EHR extraction, claims processing, de-identification
- Evaluate model performance with appropriate healthcare metrics
- Deploy analytics solutions in production environments
Course Topics
- Healthcare Data Sources: EHR, Claims, Registry, IoT
- Predictive Modeling: Regression, Random Forest, Gradient Boosting
- Clinical NLP: Extracting Insights from Unstructured Notes
- Time Series Analysis: Patient Monitoring, Demand Forecasting
- Prescriptive Analytics: Optimization, Simulation
- Model Governance: Validation, Bias, Fairness, Explainability
- Implementation: APIs, Dashboards, Clinical Decision Support
Required Textbooks
- Hastie, Tibshirani & Friedman: Elements of Statistical Learning
- Healthcare Analytics Case Studies (provided)
Evaluation Methods
Analytics Project (40%), Lab Assignments (30%), Exam (20%), Presentation (10%)