From BI to AI Prediction: The Distance Between Influenza and Death
Chi Mei Hospital
Influenza is a common, complicated seasonal disease with rapid changes among the elderly. It could lead to hospitalization, sepsis, or even death. Therefore, it is very important to predict the prognosis of this disease as early as possible. Based on the BI paper “Geriatric Influenza Death score” published in 2018, Chi Mei Hospital adopted supervised machine learning “random forest” algorithm with ten candidate predictors/features to analyze the medical records of Chi Mei system from 2009 to 2018 and established five AI prediction models for influenza. The model is integrated into the hospital information system (HIS) with Web Service at emergency department. It can promptly and automatically predict the prognosis for geriatric patients which serve as an important reference for physicians, patients, and their families.