Blood Glucose Level Prediction: Advanced Deep-Ensemble Learning Approach

• Published in: IEEE journal of biomedical and health informatics 2022-06, Vol.26 (6), p.2758-2769
• Main Authors: Nemat, Hoda, Khadem, Heydar, Eissa, Mohammad R., Elliott, Jackie, Benaissa, Mohammed
• Format: Article
• Language: English
• Subjects: Algorithms; Benchmarks; Biological system modeling; Blood; Blood glucose level; Data models; Deep learning; Diabetes mellitus; Ensemble learning; Forecasting; Glucose; Machine learning; meta-learning; Predictive models; Regression analysis; Task analysis; Time series analysis; time series forecasting
• ISSN: 2168-2194; 2168-2208
• DOI: 10.1109/JBHI.2022.3144870

Optimal and sustainable control of blood glucose levels (BGLs) is the aim of type-1 diabetes management. The automated prediction of BGL using machine learning (ML) algorithms is considered as a promising tool that can support this aim. In this context, this paper proposes new advanced ML architectures to predict BGL leveraging deep learning and ensemble learning. The deep-ensemble models are developed with novel meta-learning approaches, where the feasibility of changing the dimension of a univariate time series forecasting task is investigated. The models are evaluated regression-wise and clinical-wise. The performance of the proposed ensemble models are compared with benchmark non-ensemble models. The results show the superior performance of the developed ensemble models over developed non-ensemble benchmark models and also show the efficacy of the proposed meta-learning approaches.