# Glycemic-Aware and Architecture-Agnostic Training Framework for Blood Glucose Forecasting in Type 1 Diabetes

arXiv: https://arxiv.org/abs/2502.14183 · HTML: https://mlsanity.com/paper/2502.14183/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 52 references that are in the map.
- First posted on arXiv 20 Feb 2025.

## Builds on (most cited first, 8 of 8)

- [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting](https://mlsanity.com/paper/2012.07436/) (2020)
- [TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis](https://mlsanity.com/paper/2210.02186/) (2022)
- [Convolutional Recurrent Neural Networks for Glucose Prediction](https://mlsanity.com/paper/1807.03043/) (2018)
- [Gluformer: Transformer-Based Personalized Glucose Forecasting with Uncertainty Quantification](https://mlsanity.com/paper/2209.04526/) (2022)
- [Predicting Blood Glucose with an LSTM and Bi-LSTM Based Deep Neural Network](https://mlsanity.com/paper/1809.03817/) (2018)
- [Integration of Clinical Criteria into the Training of Deep Models: Application to Glucose Prediction for Diabetic People](https://mlsanity.com/paper/2009.10514/) (2020)
- [Hybrid Self-Attention NEAT: A novel evolutionary approach to improve the NEAT algorithm](https://mlsanity.com/paper/2112.03670/) (2021)
- [Machine Learning based prediction of Glucose Levels in Type 1 Diabetes Patients with the use of Continuous Glucose Monitoring Data](https://mlsanity.com/paper/2302.12856/) (2023)

## Built on by (most cited first, 5 of 5)

- [NutriGen: Personalized Meal Plan Generator Leveraging Large Language Models to Enhance Dietary and Nutritional Adherence](https://mlsanity.com/paper/2502.20601/) (2025)
- [AZT1D: A Real-World Dataset for Type 1 Diabetes](https://mlsanity.com/paper/2506.14789/) (2025)
- [GlyTwin: Digital Twin for Glucose Control in Type 1 Diabetes Through Optimal Behavioral Modifications Using Patient-Centric Counterfactuals](https://mlsanity.com/paper/2504.09846/) (2025)
- [GUIDE: Reinforcement Learning for Behavioral Action Support in Type 1 Diabetes](https://mlsanity.com/paper/2604.00385/) (2026)
- [MetaPlate: Counterfactual-Guided RAG-LLM Tool for Personalized Food Recommendation and Hyperglycemia Prevention](https://mlsanity.com/paper/2606.10120/) (2026)

## Abstract (arXiv)

Managing Type 1 Diabetes (T1D) demands constant vigilance as individuals strive to regulate their blood glucose levels and avoid dysglycemia, including hyperglycemia and hypoglycemia. Despite advances in automated insulin delivery (AID) systems, achieving optimal glycemic control remains challenging. These systems integrate data from wearable devices such as insulin pumps and continuous glucose monitors (CGMs), helping reduce variability and improve time in range. However, they often fail to prevent dysglycemia due to limitations in prediction algorithms that cannot accurately anticipate glycemic excursions. This limitation highlights the need for more advanced glucose forecasting methods. To address this need, we introduce GLIMMER (Glucose Level Indicator Model with Modified Error Rate), a modular and architecture-agnostic training framework for glucose forecasting. GLIMMER combines structured preprocessing, a region-aware loss formulation, and genetic algorithm-based weight optimization to emphasize prediction accuracy in dysglycemic regions. We evaluate GLIMMER using two datasets: the publicly available OhioT1DM dataset and a newly collected AZT1D dataset consisting of data from 25 individuals with T1D. Our analyses demonstrate that GLIMMER consistently improves forecasting performance across baseline architectures, reducing RMSE and MAE by up to 24.6% and 29.6%, respectively. Additionally, GLIMMER achieves a recall of 98.4% and an F1-score of 86.8% for dysglycemia prediction, highlighting strong performance in clinically high-risk regions. Compared with state-of-the-art models containing millions of parameters-such as TimesNet (18.7M), BG-BERT (2.1M), and Gluformer (11.2M)-GLIMMER attains comparable accuracy while using only 10K parameters, demonstrating its efficiency as a lightweight and architecture-agnostic solution for glycemic forecasting.