Portfolio

Predicting ICU Readmissions

A regional hospital network was seeing high 30-day ICU readmission rates with no early warning system in place. I built a predictive model trained on 3 years of EHR data that flagged high-risk patients at discharge with 84 percent accuracy, giving clinical staff a 48-hour window to intervene. Readmission rates…

Chronic Disease Risk Scoring Model

A public health organisation needed a scalable way to identify patients at high risk of developing Type 2 diabetes across a population of more than 200,000. I built a risk scoring model using demographic, lifestyle, and clinical variables that correctly identified 78 percent of high-risk individuals, enabling targeted early intervention…