DATA WAREHOUSE FOR MINING AND SIMULATING MEDICAL RECORDS

Abstract

Nowadays, patient’s data required to make informed medical decisions are trapped within fragmented and disparate clinical and administrative systems that are not properly integrated or fully utilized. Therefore, there is a growing need in the healthcare sector to store and organize sizeable record of patients to assist the healthcare sector in making informed decision. This study has been able to design and simulate the framework by developing a web application that integrates the front-end, middle-end and the back-end together as a single system using patient records from three different tropical diseases; Malaria, Measles and Typhoid Fever as the operational data source for populating the Data Warehouse. The application can be implemented in any hospital in other to aid quick decision making, most especially in certain cases for community diagnosing. Previous studies only focused on specific diseases and also visualization of clinical work flow. However, this research provides an architecture for designing a clinical data warehouse that is not limited to a single disease and will operate as a distributed system. Periodic update on a daily basis is possible because contemporary technologies have narrowed the gap between updates  which enable organizations to have a “real time” data warehouse which can be analysed using an OLAP Server. In conclusion, the study demonstrates how data can be incorporated from diverse desperate heterogeneous clinical data stores into a single data warehouse for mining and analysis purpose to aid medical practitioners and decision makers in decision-making. The study has also been able to develop data mining, reporting and analysis tool (GUI) where users can interact with the system to get a speedy and timely information needed for the clinical decision making and community diagnosing.