Predictive Data Mining and pattern recognition in the medical sector: Implementation and experience

Large enterprises essentially require a reliable infrastructure that can store, retrieve and analyze the massive volumes of high dimensional data. Knowledge Discovery in Databases (KDD) through Data Mining (DM) presents a powerful tool for storing and retrieving data in a manner that optimizes performance as well as resources. This work presents the application of Time Series Data Mining Algorithm to Hospital Management Information System (HMIS) in a public sector hospital in Pakistan. Public sector hospitals in Pakistan, apart from being typically overcrowded are severely limited by lack of resources. This research focuses on the cost effective application of KDD alongside correctly predicting disease patterns, hospital admittance rate and patient turn out. The results of this work not only bring about an improvement in management for this hospital but also provide a model for other health care facilities in the developing world.