# Deep Learning-Based Hypoglycemia Classification Across Multiple Prediction Horizons

arXiv: https://arxiv.org/abs/2504.00009 · HTML: https://mlsanity.com/paper/2504.00009/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 1 of its 23 references that are in the map.
- First posted on arXiv 25 Mar 2025.

## Builds on (most cited first, 1 of 1)

- [Metrics for Multi-Class Classification: an Overview](https://mlsanity.com/paper/2008.05756/) (2020)

## Built on by (most cited first, 3 of 3)

- [Presenting DiaData for Research on Type 1 Diabetes](https://mlsanity.com/paper/2508.09160/) (2025)
- [Benchmarking ResNet for Short-Term Hypoglycemia Classification with DiaData](https://mlsanity.com/paper/2511.02849/) (2025)
- [Impact of Age Specialized Models for Hypoglycemia Classification](https://mlsanity.com/paper/2604.23732/) (2026)

## Abstract (arXiv)

Type 1 diabetes (T1D) management can be significantly enhanced through the use of predictive machine learning (ML) algorithms, which can mitigate the risk of adverse events like hypoglycemia. Hypoglycemia, characterized by blood glucose levels below 70 mg/dL, is a life-threatening condition typically caused by excessive insulin administration, missed meals, or physical activity. Its asymptomatic nature impedes timely intervention, making ML models crucial for early detection. This study integrates short- (up to 2h) and long-term (up to 24h) prediction horizons (PHs) within a single classification model to enhance decision support. The predicted times are 5-15 min, 15-30 min, 30 min-1h, 1-2h, 2-4h, 4-8h, 8-12h, and 12-24h before hypoglycemia. In addition, a simplified model classifying up to 4h before hypoglycemia is compared. We trained ResNet and LSTM models on glucose levels, insulin doses, and acceleration data. The results demonstrate the superiority of the LSTM models when classifying nine classes. In particular, subject-specific models yielded better performance but achieved high recall only for classes 0, 1, and 2 with 98%, 72%, and 50%, respectively. A population-based six-class model improved the results with at least 60% of events detected. In contrast, longer PHs remain challenging with the current approach and may be considered with different models.