Meal optimisation to control glucose levels in people with diabetes

In diabetes, guidance of diet and exercise regime is often based on sub-optimal one size fits all approaches, so, many receive acceptable guidance, but none receive optimal guidance. We can improve guidance using digital twins capturing all relevant dynamics to personalise care. Incorporating wearable device data, this project aims to develop a patient-specific digital twin model of glucose and insulin dynamics that accounts for patient meal intake and exercise. By fitting the model to patient data every time the patient eats a meal we can deduce trends in patient condition vs meal type and timing that can be used to predict forward in time for specific meal plans. The pipeline we develop will be used by the MBIE Soul Machines AI group to determine optimal meal type and timing for the specific patient and a digital health navigator will guide the patient towards following this meal plan.

• Bachelors’ degree in Engineering, Physics, Computer Science or related discipline
• Strong programming skills
• Strong computational modelling skills
• Experience with machine learning
• Strong written and verbal communication skills
• Ability to work independently and as part of a team
  

MBIE Soul Machines Grant

Page expires: 2 January 2025