Towards Affordable, Non-Invasive Real-Time Hypoglycemia Detection Using Wearable Sensor Signals

TL;DR

This work addresses the need for affordable, non-invasive real-time hypoglycemia detection when continuous glucose monitoring is inaccessible. It builds a multimodal framework that exploits galvanic skin response and heart rate signals from wearables, with an end-to-end pipeline for preprocessing, windowing, feature extraction, and both early and late fusion using CNNs, LSTMs, GRUs, and TCNs. On the OhioT1DM 2018 dataset, multimodal fusion achieves the strongest performance, with improved recall and AUC over single-modality baselines, and ablation confirms complementary contributions of GSR and HR. The findings demonstrate that real-time hypoglycemia detection can be realized with inexpensive, non-invasive sensors, offering a practical pathway for deployment in low-resource settings and underserved communities.

Abstract

Accurately detecting hypoglycemia without invasive glucose sensors remains a critical challenge in diabetes management, particularly in regions where continuous glucose monitoring (CGM) is prohibitively expensive or clinically inaccessible. This extended study introduces a comprehensive, multimodal physiological framework for non-invasive hypoglycemia detection using wearable sensor signals. Unlike prior work limited to single-signal analysis, this chapter evaluates three physiological modalities, galvanic skin response (GSR), heart rate (HR), and their combined fusion, using the OhioT1DM 2018 dataset. We develop an end-to-end pipeline that integrates advanced preprocessing, temporal windowing, handcrafted and sequence-based feature extraction, early and late fusion strategies, and a broad spectrum of machine learning and deep temporal models, including CNNs, LSTMs, GRUs, and TCNs. Our results demonstrate that physiological signals exhibit distinct autonomic patterns preceding hypoglycemia and that combining GSR with HR consistently enhances detection sensitivity and stability compared to single-signal models. Multimodal deep learning architectures achieve the most reliable performance, particularly in recall, the most clinically urgent metric. Ablation studies further highlight the complementary contributions of each modality, strengthening the case for affordable, sensor-based glycemic monitoring. The findings show that real-time hypoglycemia detection is achievable using only inexpensive, non-invasive wearable sensors, offering a pathway toward accessible glucose monitoring in underserved communities and low-resource healthcare environments.

Figures (3)

• Figure 1: End-to-End Machine Learning Workflow for Glucose State Prediction: This flowchart outlines the complete process from data acquisition (OhioT1DM) to model comparison, highlighting preprocessing, training, imbalance handling, and performance evaluation steps.
• Figure 2: Raw GSR signal from the OhioT1DM dataset, showing tonic shifts and phasic peaks linked to autonomic activation.
• Figure 3: Multimodal wearable physiological signals used in this study