Praesiidium: the Horizon project on prediabetes risk enters its third year of activities

The annual meeting of PRAESIIDIUM, the Horizon project dedicated to developing artificial intelligence algorithms (physics-informed machine learning) to predict a person’s risk of developing prediabetes, was held in Lugano on 13 and 14 February 2025, hosted by the SUPSI Partner.

PREAESIIDIUM (Physics-informed Machine Learning-based Prediction and Reversion of Impaired Fasting Glucose Management) is funded by the European Commission under the Horizon Europe-Cluster Health programme, and is coordinated by Spindox Labs, the R&D centre of the Spindox Group.

In 2025, the final year of work, dedicated clinical studies (run by clinical partners in Austria and Latvia), consolidation of risk prediction models, exploitation analysis of results and preparation of the demo for the final events will be completed.

Early detection of this risk can prevent the development of established diabetes through the adoption of a healthy lifestyle with adequate diet and exercise.

PRAESIIDIUM will develop a tool based on artificial intelligence coupled with integrated multi-scale and multi-organ mathematical equations for real-time prediabetes risk prediction of an individual.

The prediction algorithm will make use of a rich set of training information derived from previous clinical data, the individual’s family history and a pilot study that will test wearable sensors providing glucose, bioimpedance and heart rate monitoring.

The PRAESIIDIUM platform will be made available to healthcare professionals and patients to facilitate data entry and query of results and will be linked to common wearable sensors to monitor physical activity.