Using intelligent data sources to monitor unusual behaviors in individual's health data
Date of this Version
E-health data management is characterized by high pressure and timely access. Accessing patient data requires that all services and objects are connected to make data from different health care sources available. Clinical data warehouses in this context facilitate the analysis, consolidation and access of the data obtained in the patient care process to improve the quality of decision making. There is a need to capture data events as soon as possible to decrease data analysis latency and maximize the value of information. Given the volume of data that may be generated, some sources can be improved by intelligent local processing and filtering of data for selective reporting. This paper proposes the design of a real time adaptive framework that covers the process of predicting, responding and monitoring unusual behaviors in patient data in a data warehouse environment.
This document has been peer reviewed.