Diabetes:
predictive modelling of glycaemic status

Short-term prediction of glycaemic dynamics is essential to improve Diabetes self-management. GK will provide a personalized, adaptive, real-time data driven computational solution based on data federation in the Healthcare Space, identifying the different modes of the underlying glucose metabolism and eventually prevention, of hypoglycaemic events. Advanced GK “things” will collect clinical data at home such as bio- and physiological signals (i.e. blood glucose concentration data or continuous glucose monitoring data, galvanic skin response, heart rate variability) combining them with adaptive machine-learning regression models.

Target population:

Although all sites plan on including patients 65+ years with diabetes or poor metabolic control and associated comorbidities, the specific target population varies from site to site.

Key enabling technologies:

  • Wearables/medical devices intended to monitor and track key variables including: glucose levels; physical activity; sleep pattern; blood pressure; weight and body composition; and adherence to treatment with electronic pill-boxes. 
  • App for smartphone or tablet, enabling self-management, healthy lifestyle promotion, and regular follow up of patients through their interaction with a digital coach or chat-bot, including the possibility of answering questionnaires; screening cognitive, behavioural and mood status; and, monitoring drug intake and fostering adherence. 
  • Professional’s online platform/dashboard, providing overview of alarm signs and/or relevant information. • Raspberry Pi Gateway (Greece). 
  • Homebound educational support based on a social robot (Puglia, optional).
  • Continuous Glucose Monitoring system.