How can artificial intelligence help researchers understand complex disease processes and identify previously hidden patterns? That was the focus when Fang Fang, professor at the Institute of Environmental Medicine, Karolinska Institutet, was interviewed for the internationally renowned medical journal JAMA's podcast series JAMA+ AI Conversations.
In the interview, Fang Fang discusses how AI-based methods can be used to analyze very large and complex datasets in epidemiological research, particularly in studies of neurodegenerative diseases such as ALS, Parkinson's disease, and dementia.
A central theme is the unique potential of the Nordic health registers. Thanks to personal identity numbers and long-term registry data, researchers can follow individuals' health across much of their lives and combine information on healthcare, medications, family relations, and socioeconomic factors. According to Fang Fang , this makes the Nordic countries especially well suited for AI-driven research.
She also highlights that AI is particularly valuable for handling so-called multimodal data - large amounts of information from many different sources simultaneously - something traditional statistical models often struggle to manage efficiently. AI can therefore help researchers uncover previously invisible disease patterns and map disease trajectories over time.
At the same time, Fang Fang emphasizes the importance of scientific caution. The predictive power of AI must be combined with epidemiological and biological understanding to ensure that findings are both interpretable and scientifically sound.
The interview is part of JAMA+ AI Conversations, a series featuring researchers and experts discussing the rapidly growing use of AI in medicine and health sciences.
You can listen to the interview here:
