Introduction

To address the challenge of obesity, artificial intelligence (AI) is emerging as a powerful tool for developing personalised health interventions. To mark the International Day of Scientific Culture and understand how these technologies may shape the daily lives of people living with obesity, we hear from Prof. Dr. Stavroula Mougiakakou, head of the AI in Health and Nutrition Laboratory at the University of Bern. In this interview, Prof. Dr. Mougiakakou shares insights on how AI is shaping the future of healthcare by uncovering complex causal relationships between genetics, behaviour, and the environment, and how this innovative approach could revolutionise obesity prevention and treatment.

What is your role in BETTER4U?

The AI in Health and Nutrition Laboratory at the University of Bern is leading efforts in designing and developing an AI model aimed at identifying the key personal factors that most significantly influence an individual’s weight, while also providing tailored recommendations for optimizing weight management. This model will be built using knowledge contributed by domain experts involved in the BETTER4U project, as well as insights from analysing retrospective datasets of 1 million individuals and data collected from a prospective pilot study as part of the project.

The AI model will define and quantify causal relationships between various factors linked to weight gain and obesity, including genetic, biological/anthropometric, behavioural/lifestyle, and environmental variables. By establishing these causal connections, the model will enable predictions on the outcomes of different interventions and offer counterfactual estimations. This will empower healthcare professionals to develop personalised, evidence-based interventions.

What is AI, and why is it relevant in health-related projects like BETTER4U?

AI refers to the science and engineering of enabling computers and machines to learn from experience, adapt to changing environments, and perform tasks that typically require human intelligence. This includes techniques like machine learning, computer vision, natural language processing, and deep learning, which are applied in various fields, including healthcare.

In health-related projects like BETTER4U, AI is highly relevant because it allows us to analyse vast amounts of heterogeneous data and uncover complex patterns that would be challenging for humans to identify. For example, in the BETTER4U project, AI will help determine the key factors influencing individual weight management by analysing data from multiple sources, such as retrospective datasets and pilot studies. AI also enables the creation of personalised recommendations for patients based on their unique health and lifestyle profiles. This personalised approach improves the accuracy of diagnoses, predictions, and treatments, leading to better health outcomes.

Furthermore, AI can model causal relationships between variables related to obesity, such as genetics, behaviour, and the environment, allowing healthcare professionals to make evidence-based decisions and optimise interventions. By supporting healthcare professionals in this way, AI helps advance the goal of personalised medicine, which tailors treatments to individual needs, ultimately improving care and resource use.

In summary, AI’s capabilities in knowledge representation, reasoning, and its ability to process and learn from vast, diverse datasets make it an invaluable tool for tackling complex global health challenges, such as those addressed in the BETTER4U project. Additionally, AI has the potential to promote equity in healthcare by providing personalised care regardless of socioeconomic background, though careful attention must be paid to mitigating bias in algorithms and data. Moreover, AI solutions can be designed for implementation in low-resource settings, ensuring that advanced health technologies are accessible and beneficial to all, particularly underserved populations.

What role does AI play in the BETTER4U project specifically, and how does it help address the global obesity challenge?

In the BETTER4U project, AI plays a crucial role in identifying key factors influencing individual weight and providing personalised recommendations for weight management. By analysing large datasets and identifying causal relationships, AI helps tailor interventions to each person’s unique profile, moving beyond one-size-fits-all solutions.

Additionally, AI promotes equitable healthcare by enabling personalised, scalable interventions that can be applied in low-resource settings. By addressing bias and ensuring fairness, AI contributes to a more inclusive approach to the global obesity challenge.

What key benefits do you anticipate from using AI-driven solutions in the BETTER4U project compared to traditional weight-loss methods?

AI contributes to greater personalisation and precision by enabling tailored interventions based on an individual’s genetic, behavioural, biological/anthropometric, cultural/religious, and environmental factors, moving beyond one-size-fits-all approaches. The real advantage, however, lies in AI’s ability to process vast amounts of data, uncovering insights that traditional methods might miss. It can predict intervention outcomes and optimise treatment plans dynamically, allowing for continuous monitoring and adjustment, thereby enhancing healthcare professionals’ ability to provide responsive, personalised care. AI-driven solutions are scalable and can be applied even in low-resource settings, where healthcare experts may not be available. This extends personalised, data-driven strategies to broader populations, ensuring that individuals receive tailored support even in underserved areas. By complementing the expertise of healthcare professionals, AI empowers them to deliver more effective, individualised care and helps address the global challenge of obesity more efficiently.

What challenges have you encountered, particularly in addressing a complex issue like obesity? Are there specific ethical or privacy-related concerns?

In addressing a complex issue like obesity, we’ve encountered several challenges. One significant technical challenge lies in handling heterogeneous and distributed datasets from different sources, which often vary in format, quality, and completeness. This complexity makes data integration and analysis more difficult.

Additionally, because BETTER4U is a multidisciplinary project, not all participants have the same level of literacy in each domain involved—be it AI, healthcare, or data science. This can lead to communication gaps and misalignment of expectations, requiring extra effort to ensure all stakeholders are on the same page.

Ethical and privacy concerns also play a crucial role, particularly when handling sensitive data. Ensuring compliance with data privacy regulations, maintaining transparency, and addressing potential biases in AI models are top priorities.

It’s important to note that we’re still in the early stages of the project, around the end of the first year. Therefore, this list of challenges is not yet exhaustive. As we move forward, we will refine our processes and streamline activities to address these issues effectively and meet the project’s objectives. Despite the hurdles, we remain very positive and enthusiastic that the challenges will be solved as we progress.

How do you envision health professionals applying the results of BETTER4U in real-world settings to personalise interventions for patients?

In an ideal scenario, health professionals will leverage AI-driven insights from BETTER4U to personalize interventions based on each patient’s unique genetic, behavioural, and environmental factors. By identifying the most influential factors affecting weight, experts can create more precise, tailored treatment plans. This will enable them to predict the outcomes of interventions—such as diet, exercise, or medication—adjust plans dynamically, and optimize results in real time. By integrating these insights into digital health platforms, professionals can enhance their expertise with data-driven recommendations, ultimately delivering more effective, individualised care. That said, we are still in the early stages of the project, and many aspects of this vision may need to be adjusted based on feedback from participants and healthcare professionals as we progress.

Looking ahead, what future directions do you see for the BETTER4U project, and how do you think AI’s role in addressing obesity will evolve?

Looking ahead, the BETTER4U project is poised to drive a paradigm shift in healthcare, transitioning from treating symptoms and diseases to preventing their onset through effective self-management and a deeper understanding of genetic predispositions, environmental influences, and lifestyle factors. As AI continues to advance, it will play an increasingly vital role in enabling individuals and healthcare providers to anticipate and manage health risks more proactively.

In the future, AI’s role in addressing obesity will likely extend beyond clinical settings, becoming seamlessly integrated into everyday life through digital health platforms and wearable devices. This will empower individuals to monitor their health in real time and receive personalised guidance, enhancing self-management and promoting long-term well-being. BETTER4U has the potential to lead this transformation, shaping a future where prevention, rather than treatment, becomes the cornerstone of healthcare.

As we mark the International Day of Scientific Culture, what would you like the public to understand about the role of AI and scientific innovation in addressing global health issues like obesity?

As we mark the International Day of Scientific Culture, it’s important for the general population to understand that AI is a man-made tool—its true potential depends on how we choose to use it for the benefit of humanity. In addressing global health issues like obesity, AI offers powerful capabilities to personalise care, predict health risks, and optimise interventions. However, to truly harness its power, close collaboration between all involved parties—scientists, healthcare professionals, policymakers, and society at large—is essential.

“AI cannot operate in isolation; it requires thoughtful guidance, ethical considerations, and the active participation of everyone. By working together, we can ensure that AI-driven innovations serve humanity, enhancing health outcomes and making equitable, accessible healthcare a reality for all.”

Prof. Dr. Stavroula Mougiakakou, University of Bern