Objective
The MELISSA project aims to provide a clinically validated, effective, trustworthy, and cost-efficient artificial intelligence (AI)-based digital diabetes management solution to support both health care providers (HCPs) and insulin-treated patients with diabetes (PwD) in their daily routine with personalised treatment and care recommendations. The solution is independent of the used glucose monitoring devices and is based on the combined use of already prototyped advanced AI-approaches and innovative tools for quantification of lifestyle and behavioural factors, taking into consideration sex/gender aspects, age, and socio-economic parameters related to the development of diabetes. More specifically, core element of the project is the daily insulin treatment adjustment to ensure glucose control using an already introduced self-learning approach based on reinforcement learning. The approach is data-driven, real-time and of low computational cost and allows daily adjustment of the insulin infusion profile, on the basis of the fluctuations in the patient?s glucose. The approach takes into consideration patients treatment-related (glucose, insulin, and carbohydrate intake) and conceptual information and during the project will be further extended and optimized to include additional lifestyle and behavioural parameters. Furthermore, tools for assessing the risk of short- and long-term complications will assist HCPs in reaching better decisions on adjustments to the treatment schemes. To meet the objectives the consortium brings together partners active in the fields of diabetes (PwD and HCPs), diabetes technology, AI, behavioral sciences, ethics in AI, regulatory affairs, healthcare economics and clinical trials to further co-create, clinically develop, optimize, and clinically validate an AI-based solution for more effective and cost-efficient diabetes management through personalised treatment and care.
Fields of science
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
- natural sciencescomputer and information sciencesartificial intelligencemachine learningreinforcement learning
- medical and health sciencesclinical medicineendocrinologydiabetes
- social scienceseconomics and businesseconomicshealth economicseconomic impact of epidemics
- natural sciencesbiological sciencesbiochemistrybiomoleculescarbohydrates
Keywords
Programme(s)
Funding Scheme
HORIZON-RIA - HORIZON Research and Innovation ActionsCoordinator
6200 MD Maastricht
Netherlands