A newly developed tool detects low blood sugar levels based on driving behavior and head/gaze motion.
Image Credit: Eviart/Shutterstock.com
One of the most serious effects of diabetes is low blood sugar, or hypoglycemia, which increases the risk of cognitively taxing activities requiring sophisticated motor skills, like operating a vehicle. The low availability, high cost, invasiveness, and diagnostic delay of current tools limit their utility in detecting hypoglycemia.
A novel method for identifying hypoglycemia while driving is offered by a recent study that was published in the journal NEJM AI. Scientists from LMU conducted the research in cooperation with colleagues from the University of St. Gallen, ETH Zurich, and the University Hospital of Bern (Inselspital).
Thirty diabetics participated in the study, providing data while operating a real car. Data was recorded for each patient twice: once when their blood sugar was normal and again when they were hypoglycemic.
To this end, medical personnel in the car purposefully caused each patient to become hypoglycemic. Driving signals like vehicle speed and head/gaze motion data, like the pace of eye movements, were among the data that were gathered.
This study not only showcases the potential for AI to improve individual health outcomes but also its role in improving safety on public roads.
Stefan Feuerriegel, Head and Project Partner, Institute of Artificial Intelligence (AI) in Management, Ludwig-Maximilians-Universität München
The researchers then created a brand-new machine learning (ML) model that can identify hypoglycemic episodes automatically and with high accuracy based only on regularly gathered driving and head/gaze motion data.
This technology could serve as an early warning system in cars and enable drivers to take necessary precautions before hypoglycemic symptoms impair their ability to drive safely.
Simon Schallmoser, Doctoral Candidate and Contributing Researcher, Institute of AI, Management, Ludwig-Maximilians-Universität München
The newly developed ML model also performed well when only head/gaze motion data was used, which is crucial for future self-driving cars.
This study not only showcases the potential for AI to improve individual health outcomes but also its role in improving safety on public roads.
Stefan Feuerriegel, Head and Project Partner, Institute of Artificial Intelligence (AI) in Management, Ludwig-Maximilians-Universität München
Journal Reference:
Lehmann, V., et.al (2024) Machine Learning to Infer a Health State Using Biomedical Signals – Detection of Hypoglycemia in People with Diabetes while Driving Real Cars. Journal of Medicine AI. doi.org/10.1056/Aioa2300013.
Source: https://www.lmu.de/en/