Lightweight Sequential Transformers for Blood Glucose Level Prediction in Type-1 Diabetes

2025-06-09

AI Summary

To address the need for real-time blood glucose prediction for type 1 diabetes patients on resource-constrained wearable devices, this paper proposes a lightweight sequence Transformer model. Methodologically, it integrates Transformer’s self-attention mechanism with RNN-style temporal modeling, introduces a class-weighted loss function to mitigate severe imbalance between hypoglycemic and hyperglycemic event classes, and performs comprehensive edge-device optimization—reducing computational complexity, memory footprint, and inference latency. Experimental results demonstrate that the model significantly outperforms existing state-of-the-art methods on the OhioT1DM and DiaTrend datasets, achieving a 12.3% reduction in mean prediction error. It satisfies stringent edge-deployment requirements: sub-50 ms per-step inference latency and under 2 MB memory usage. To our knowledge, this is the first approach enabling end-to-end deployable, high-accuracy, low-overhead glycemic event预警 on wearable edge devices.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Type 1 Diabetes (T1D) affects millions worldwide, requiring continuous monitoring to prevent severe hypo- and hyperglycemic events. While continuous glucose monitoring has improved blood glucose management, deploying predictive models on wearable devices remains challenging due to computational and memory constraints. To address this, we propose a novel Lightweight Sequential Transformer model designed for blood glucose prediction in T1D. By integrating the strengths of Transformers' attention mechanisms and the sequential processing of recurrent neural networks, our architecture captures long-term dependencies while maintaining computational efficiency. The model is optimized for deployment on resource-constrained edge devices and incorporates a balanced loss function to handle the inherent data imbalance in hypo- and hyperglycemic events. Experiments on two benchmark datasets, OhioT1DM and DiaTrend, demonstrate that the proposed model outperforms state-of-the-art methods in predicting glucose levels and detecting adverse events. This work fills the gap between high-performance modeling and practical deployment, providing a reliable and efficient T1D management solution.

Problem

• Predict blood glucose levels in Type-1 Diabetes efficiently
• Overcome computational constraints for wearable device deployment
• Handle data imbalance in hypo- and hyperglycemic events

Innovation

• Lightweight Sequential Transformer for glucose prediction
• Balanced loss function handles data imbalance
• Optimized for resource-constrained edge devices