Seasonal influenza affects 5-10% of adults worldwide annually. With the increased use of electronic medical records, large diagnostic and laboratory datasets are available in real-time and are a promising supplement to traditional surveillance activities. Diagnostic datasets of International Classification of Diseases (ICD) codes by themselves are not sensitive enough for tracking influenza incidence as influenza patients are often coded as influenza-like illnesses.In this study, we explore the application of machine learning methods using diagnostic and virologic data to get an estimate of influenza incidence.
This is my paper for final project for Fall 2017 course: P9120 Topics in Statistical Learning & Data Mining taught by Min Qian.