Stavroula Mougiakakou, head of the research group for Artificial Intelligence in Health and Nutrition at the ARTORG Center in Bern, conducts applied fundamental research and with her team aims to close the gap between theory and practice.
Interview: Michael Gasser / English Version: Monika Kugemann / Photos: Dres Hubacher
Ms. Mougiakakou, how would you describe your main field of research and your scientific approach?
Since my studies, my research has focused on artificial intelligence (AI) and for several years I have been concentrating on its application in nutrition and health. In my lab, we are conducting basic, applied and translational research in AI; our objective is to bridge the gap between theory and practice – in both clinical and real-life settings. Our interdisciplinary team analyses the needs, shapes the technology to meet real challenges and provides users with relevant data related to the diagnosis and management of acute and chronic disease. Our latest focus, for instance, is on the diagnosis, management, and prognosis of lung diseases.
What made you decide to pursue a career in Biomedical Engineering?
In 1996, during my diploma thesis, I already had the opportunity to discuss the best approaches to applying engineering in medicine with physicians and engineers as part of my studies in electrical engineering and information technology in Athens. I attended an international conference on biomedical engineering, where I met other researchers who had already succeeded in applying biomedical engineering to help patients. With this incentive, I decided to pursue my PhD in the field of AI in Medicine, an area which was rather new then. Since then, I have been working in the field of AI for individualized diagnosis and treatment – not only with medication, but also with lifestyle parameters such as food intake.
Where do you find ideas and inspiration for upcoming research projects?
Let me tell you about our new project: This is funded by the European Commission and the State Secretariat for Education, Research, and Innovation (SERI). Its main objective is to provide a clinically validated, efficient and cost-effective AI-based solution for the digital management of diabetes to support insulin-treated patients with diabetes and their health care providers, by providing personalized recommendations for treatment and care.
Launched in June 2022, the project is the result of innovations developed by my group over the past 14 years. Now, for the first time, we have the opportunity not only to integrate the technology, but also to deploy it in five clinical facilities – in four European countries. The project itself is based on the close collaboration of my team with an industrial partner and a number of physicians across Europe. Especially, with the members of my team we share the same passion. The path to our goal resembles a long, steep, and often rocky road, with small successes along the way. These motivate us to push our research further.
What distinguishes your field of research from others?
This is a highly interdisciplinary field – from the initial ideas to the practical implementation. If we are to successfully develop an idea for a system and translate this for the benefit of the society, we must meet a variety of different challenges, and work closely with experts in other areas, including engineers, computer scientists, health care professionals, patients, and experts in ethics in AI for medicine.
The path resembles a long, steep, and often rocky road, with small successes along the way.
Where do you see the biggest hurdles for your work and your research?
The entire field of AI in medicine is very competitive at the national and international levels, which is per se positive, but on the other hand, can lead to enormous pressure, especially when there is clear need for rigorous scientific testing that can assure approaches, tools and processes that can benefit users. There is always a need for both-cutting edge technologies and in parallel to present clinically validated approaches. ARTORG’s approach is to integrate engineering technology and medicine. This leads to clinically embedded development and is extremely helpful in closing the substantial gap between the research lab and the end-user – either patients or health care professionals.
Which aspects of your research are you particularly proud of?
That I was among the first to apply the concept of artificial intelligence to diabetes self-management particularly in the case of insulin -dependent diabetic patients, followed by my research in AI-powered dietary assessment. We have been among the first groups worldwide that have designed and developed a complete system for the translation of food images to nutrient content and have validated this in several preclinical studies. Besides that, I feel that one of my most important accomplishments is that our PhD programs and the newly established MSc in AI in Medicine have helped to shape a generation of highly skilled, high -quality experts in AI in medicine.
What does your weekly work look like?
Generally, I spend a lot of time developing new ideas with my colleagues, competitively applying for research funds, and continuously conducting concrete research work. In addition, I supervise the research of the fellows, oversee the publication of our scientific results, and manage ongoing projects. Teaching is also an important part of my work and I lead the newly established master's program in AI and Medicine. In parallel, together with some of my colleagues, we initiated an initiative “Diversity for AI in Medicine” as part of the Center for Artificial Intelligence in Medicine - with the objective of encompassing activities in the areas of networking, outreach, mentoring and research.
Without which tool would your research not be possible?
Considering that the computation infrastructure to conduct this type of research is given, it is essential to have a research project group that works as a team. We need to share the same interest and passion for what we are doing and respect each other. We should not be afraid of failures and must persist in attempting to solve problems that sometimes even seem unsolvable.
How have you been and are you being supported in your career?
I feel very fortunate, because I have a strong international network that supported me in all the challenging moments in my career. I always received honest and constructive feedback, which is important to advance academically – in addition to the necessary excellence in science.
Does your life still leave room for interests other than research?
I am a mother of teenage twins and wife to a busy husband. Even though I love my research, my daughters and husband clearly take first place for me. It is challenging at times, but I usually manage to make time for family and enjoy as much high-quality time as possible with them.
Stavroula Mougiakakou is Associate Professor of Biomedical Engineering and heads the research group AI in Health and Nutrition at the ARTORG Center for Biomedical Engineering Research, University of Bern.Her current research interests include artificial intelligence, machine learning, machine vision and advanced data analysis. She is working on health promotion solutions in the areas of improved diagnosis, personalized treatment, and nutritional analysis.