Machine learning for non-invasive sensing of hypoglycemia while driving in people with diabetes.

Lehmann, Vera; Zueger, Thomas; Maritsch, Martin; Kraus, Mathias; Albrecht, Caroline; Bérubé, Caterina; Feuerriegel, Stefan; Wortmann, Felix; Kowatsch, Tobias; Styger, Naïma; Lagger, Sophie; Laimer, Markus; Fleisch, Elgar; Stettler, Christoph (2023). Machine learning for non-invasive sensing of hypoglycemia while driving in people with diabetes. Diabetes, obesity & metabolism, 25(6), pp. 1668-1676. Wiley 10.1111/dom.15021

Preview

Text Diabetes_Obesity_Metabolism_-_2023_-_Lehmann_-_Machine_learning_for_non_invasive_sensing_of_hypoglycemia_while_driving_in.pdf - Accepted Version Available under License Publisher holds Copyright. Download (5MB) | Preview

AIMS

Hypoglycemia is one of the most dangerous acute complications of diabetes mellitus and is associated with an increased risk of driving mishaps. Current approaches to detect hypoglycemia are limited by invasiveness, availability, costs, and technical restrictions. In this work, we developed and evaluated the concept of a non-invasive machine learning (ML) approach detecting hypoglycemia based exclusively on combined driving (CAN) and eye tracking (ET) data.

MATERIALS AND METHODS

We first developed and tested our ML approach in pronounced hypoglycemia, and, then, we applied it to mild hypoglycemia to evaluate its early warning potential. For this, we conducted two consecutive, interventional studies in individuals with type 1 diabetes mellitus. In study 1 (n=18), we collected CAN and ET data in a driving simulator during eu- and pronounced hypoglycemia (blood glucose [BG] 2.0 - 2.5 mmol L-1 ). In study 2 (n=9), we collected CAN and ET data in the same simulator but in eu- and mild hypoglycemia (BG 3.0 - 3.5 mmol L-1 ).

RESULTS

Here, we show that our ML approach detects pronounced and mild hypoglycemia with high accuracy (area under the receiver operating characteristics curve [AUROC] 0.88±0.10 and 0.83±0.11, respectively).

CONCLUSIONS

Our findings suggest that an ML approach based on CAN and ET data, exclusively, allows for detection of hypoglycemia while driving. This provides a promising concept for alternative and non-invasive detection of hypoglycemia. This article is protected by copyright. All rights reserved.

Interest & Impact

Downloads

22 since deposited on 16 Feb 2023
22 in the past 12 months

Citations

5 Citations in Web of Science ®
6 Citations in Scopus

Search

in Google Scholar™

Services

Actions (login required)

Edit item

Actions (login required)

Edit item