The overall objective of the HEADWIND project ("Design and Evaluation of a Vehicle Hypoglycemia Warning System in Diabetes") is a novel approach to improve road safety for patients with diabetes mellitus. Hypoglycemia can be a serious acute complication of diabetes mellitus treated with insulin or certain other drugs. Hypoglycemia is characterized by a reduction in concentration, a slowing down of perception and thought processes, as well as limitations in many psychomotor functions. This is particularly critical in road traffic, where rapid decision-making processes involving a large number of factors are essential. In order to reduce the increased accident risk of people with diabetes mellitus due to the risk of hypoglycaemia, the interdisciplinary and university-spanning research team will take a completely new approach, combining the immense possibilities of the rapidly developing automotive industry with innovative approaches from the field of artificial intelligence. The research team intends to detect hypoglycemia directly from the data collected from the vehicle in real time while driving.
Already today, hundreds of driving parameters are recorded during the car journey. These data will now be used and continuously analyzed by means of so-called "machine learning" to detect changes in driving behavior that indicate hypoglycemia. In a first step, the researchers will carry out tests on the driving simulator, whereby patients will be placed in hypoglycemia under medical supervision. In a next step, these investigations will be transferred to real cars on closed test tracks. A major challenge of this project, in addition to data extraction and real-time processing using complex mathematical algorithms, is the controlled induction of hypoglycemia in a moving car, an undertaking that is very complex from a logistical and medical point of view and represents a "world premiere".