Beyond Glucose-Only Assessment: Advancing Nocturnal Hypoglycemia Prediction in Children with Type 1 Diabetes

TL;DR

This study addresses nocturnal hypoglycemia prediction in children with Type 1 Diabetes by integrating wearable physiological signals with machine learning to move beyond glucose-only inputs and tackle long-horizon forecasts. It leverages an in-house pediatric dataset and transfers knowledge from the OhioT1DM adult dataset to bolster predictive performance. The results show an AUROC of 0.75 on the in-house data with a personalized glucose feature and Random Forest, which improves to 0.78 ± 0.05 through transfer learning, with reduced variability. These findings highlight the value of multi-modal physiological data and cross-domain learning for enhancing pediatric diabetes management and patient safety at night.

Abstract

The dead-in-bed syndrome describes the sudden and unexplained death of young individuals with Type 1 Diabetes (T1D) without prior long-term complications. One leading hypothesis attributes this phenomenon to nocturnal hypoglycemia (NH), a dangerous drop in blood glucose during sleep. This study aims to improve NH prediction in children with T1D by leveraging physiological data and machine learning (ML) techniques. We analyze an in-house dataset collected from 16 children with T1D, integrating physiological metrics from wearable sensors. We explore predictive performance through feature engineering, model selection, architectures, and oversampling. To address data limitations, we apply transfer learning from a publicly available adult dataset. Our results achieve an AUROC of 0.75 +- 0.21 on the in-house dataset, further improving to 0.78 +- 0.05 with transfer learning. This research moves beyond glucose-only predictions by incorporating physiological parameters, showcasing the potential of ML to enhance NH detection and improve clinical decision-making for pediatric diabetes management.

Figures (2)

• Figure 1: Two component principal component analysis (PCA) representation of the in-house dataset before and after ADASYN haibohe2008adasyn augmentation.
• Figure 2: This figure displays the distribution of the mean AUROC cross-validation scores across three different random seeds for the different feature sets across models.