Diabetes management through artificial intelligence and gamification : blood glucose prediction models and mHealth design considerations

Details

Supervisors

Macq, Benoît

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil biomédical

Abstract

This Graduation Project aimed to : 1. Highlight the diabetes management issues ; 2. Underline the importance of blood glucose prediction and empowerment ; 3. Review state-of-the-art blood glucose (BG) prediction models ; 4. Design and implement (Matlab) a nonlinear autoregressive with exogenous inputs (NARX) model for BG prediction ; 5. Suggest the development of an innovative mHealth application dedicated to diabetes self-management, including amongst others : (a) a motivational system relying on key ingredients of Gamification mechanics ; (b) an intelligent blood glucose prediction tool, able to detect and prevent the patient from hyperglycemia or hypoglycemia ; (c) an accurate, unobtrusive and non-invasive health tracking tool (e.g. multisensor armband, food barcode scanner, etc.).