Personalized precision medicine - Preventing Autoimmune diseases

A key highlight of the development work is the use of individual survival prediction (ISP) models, which move beyond traditional statistical strategies for risk determination by offering personalized insights into when a subject is likely to develop Type 1 Diabetes. The models have been trained and validated using data from The Environmental Determinants of Diabetes in the Young (TEDDY; https://teddy.epi.usf.edu/index.html) study and demonstrated the potential to optimize screening frequency and improve follow-up strategies. The predictor models have been developed for implementation on the MainlyAI platform.

AI-based models leverage individual data, including genetics (such as HLA typing), autoantibody profiles and patient history to predict risk. Unlike traditional statistical approaches, which assess risk at the population level over standardized intervals, the novel ISP enhances predictive accuracy by estimating the likelihood of disease onset at an individual level. The models offer significant potential in clinical decision-making, allowing healthcare providers to customize screening schedules and follow-up frequency on an individual level. High-risk individuals may benefit from closer, more frequent monitoring, optimizing early detection and intervention strategies.

The ASSET project, funded by the Swedish Innovation Agency (Vinnova) since 2021 and coordinated by Diamyd Medical, is focused on evaluating how AI can enhance Type 1 Diabetes screening strategies by integrating genetic data, clinical and digital biomarkers to predict disease onset with greater precision.

As AI-driven approaches continue to evolve, the ASSET consortium envisions a future where such strategies can also support the implementation of a precision medicine approach to Type 1 Diabetes, through personalized interventions and improved disease management. Further research will focus on refining predictive models and showcasing how AI could be used in facilitating earlier diagnosis and treatment selection within Type 1 Diabetes.

ASSET is a collaboration between Diamyd Medical, Mainly AI, Lund University, Sahlgrenska and Örebro University hospitals, the National Diabetes Registry and Leading Healthcare foundation.