twitter facebook

Models and simulation techniques for discovering diabetes influence factors

MOSAIC management tool

Thanks to multiple databases available from different European countries, MOSAIC has created several new prediction models. The probabilistic models’ goal is to identify people at risk of developing type 2 diabetes (or currently diabetic patients). The temporal data mining model analyzes the evolution of diabetes with the aim of predicting and preventing the appearance of complications.

The models developed have been integrated into three different MOSAIC TOOLS:

Risk factors for type 2 diabetes detection

This tool enables Public & private bodies and healthcare professionals to identify patients in risk of developing T2DM from a general population to the individual patient through the information available at Health Information Systems, focusing the completion of diagnostic tests on those patients actually in high risk, increasing success rates on tests like OGTT; and bringing to light undiagnosed T2DM and pre-diabetic states. A version of the tool will be available for citizens self-screening on the MOSAIC web or via APP.

Hospital Care Management

This tool is used by a healthcare agency/hospital head of service to understand the processes that are deployed in the health care area in a friendly way. Users can visualize summary statistics on the assisted population and, with the help of advanced data mining techniques, can explore the most frequent temporal patterns in terms of hospitalizations, complications, disease evolution, and treatment options.

Clinical Decision Support in Follow-up Visits

During follow-up visits, physicians are assisted for a better management of an individual patient. It shows a snapshot of the individual patient’s situation at the most recent encounter, and to visualize his/her whole temporal history related to interesting clinical variables, lifestyle, levels of complexity, risk scores, and medication patterns followed.

Copyright © 2013 MOSAIC project. All rights reserved.

twitter facebook