Better Quality of Life for People Living with Diabetes Through Validated Innovative Artificial Intelligence Applications: IEAI’s role in the MELISSA Project

The embedding of ethical principles in the development process of the MELISSA Projects serves to enhance the acceptance of- and trust in- the solution meant to support diabetes patients in their daily lives. On the occasion of Word Diabetes Day, marked every year on November 14th, we spoke with Mbangula Lameck Amugongo, IEAI Researcher and Member of the MELISSA Team.

How can the MELISSA Project contribute to achieving a better quality of life for people with diabetes?

The MELISSA project has the potential to create an explainable AI-based solution, the first of its kind, for diabetes care that will empower patients by personalizing their diabetes management. We aim to help patients make better decisions about their diet and physical activities. With the app, we want to help calculate the right amount of insulin that needs to be given to a patient at the right time.

How can AI transform diabetes care?

AI has the potential to “democratize” diabetes care, ensuring that quality and affordable diabetes care is available for all. However, for AI to be adopted in healthcare, patients and doctors need to trust the decisions made by AI models.

What is your role in the Project?

I am investigating technical aspects related to trustworthy AI. We intend to improve public trust in the AI solution by incorporating explainability (i.e. explain how the algorithm makes decisions to humans) into MELISSA. The aim is to ensure that intended users can trust the MELISSA solution.

Learn more about the MELISSA Project here: https://www.melissa-diabetes.eu/

At a glance: Why MELISSA is needed.

Diabetes is on the rise worldwide. In Europe alone 60 million people are living with this complex disease. Nearly 36 million people are at risk of developing diabetes if no further actions are taken. This implies that by 2030 more than 10% of the European population could be affected. The impact of such a development is immense: not only are more people at risk of diabetes complications, premature frailty and death, but also healthcare systems will be stretched to their limits due to rising costs associated with complications and the very diverse treatment of patients. Despite major advances in drug discovery and the development of innovative diabetes care over the last 20 years, not every patient can access or afford it. A growing number of people with diabetes require insulin therapy. This highly complex form of treatment greatly impacts the everyday life of people affected by diabetes and requires substantial effort and continuous commitment on their part. The introduction of AI-powered solutions for personalized treatment and care promises to become a game changer. Not only could an estimated 400,000 European lives be saved every year, it could also have a significant economic impact in annually saving €200 billion and 1.8 billion staff hours in Europe alone.