Disease predictive modeling for healthcare management system

This study attempts to perform predictive analytics for decision makers in healthcare management systems using surveillance data from multiple sources for formulating intervention programs based on the results. With the availability of big data in health from multiple sources including electronic health records, it is possible to integrate data and perform near real-time predictive analysis for incoming streams of disease incidences. We use a temporal predictive Auto-Regressive Integrated Moving Averaging model (ARIMA) in combination with a minimum size moving window to forecast the disease incidences over a data collection and integration framework. We applied our model for predictive analysis of Hepatitis C incidences in Vehari District of Punjab province in Pakistan. Model performance is evaluated based on Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The model is capable of finding trends of any disease to aid timely decision making in the healthcare management context.