Automated carbohydrate counting and machine learning could improve glycemic control in youth living with type 1 diabetes

People with type 1 diabetes (T1D) treated with a basal bolus insulin regimen need to match their insulin bolus calculation with the estimated carbohydrate content of food to maintain glucose control. We want to provide them with a more flexible approach and give them the opportunity to quickly adapt mealtime insulin using automated carbohydrate counting technology. Considering that such application does not exist yet in Quebec, we aim to develop an application that would simplify food journaling through a Quebec-specific food database and a machine learning food recognition algorithm that would automate carbohydrate counting. This research spanning over the course of 3 years is three-fold; a co-design phase will integrate the perspective of youth and healthcare professionals in the field. It will then be tested in a proof-of-concept study before larger implantation across three healthcare sites to validate its impact on glycemic control in youth with T1D treated with basal bolus insulin regimen.

Faculty Supervisor:

Anne-Sophie Brazeau

Student:

Partner:

Ikigai Développement Inc.

Discipline:

Life Sciences

Sector:

Professional, scientific and technical services

University:

McGill University

Program: