Integrating Biological-Informed Recurrent Neural Networks for Glucose-Insulin Dynamics Modeling
This work addresses the problem of patient-specific variability in glucose-insulin modeling for Artificial Pancreas systems, offering a personalized approach that could improve diabetes management, though it is incremental as it builds on existing neural network methods with biological constraints.
The study tackled the challenge of accurately modeling glucose-insulin dynamics for Type 1 Diabetes management by introducing a Biological-Informed Recurrent Neural Network (BIRNN) framework, which outperformed traditional linear models in glucose prediction accuracy and state reconstruction using a simulator.
Type 1 Diabetes (T1D) management is a complex task due to many variability factors. Artificial Pancreas (AP) systems have alleviated patient burden by automating insulin delivery through advanced control algorithms. However, the effectiveness of these systems depends on accurate modeling of glucose-insulin dynamics, which traditional mathematical models often fail to capture due to their inability to adapt to patient-specific variations. This study introduces a Biological-Informed Recurrent Neural Network (BIRNN) framework to address these limitations. The BIRNN leverages a Gated Recurrent Units (GRU) architecture augmented with physics-informed loss functions that embed physiological constraints, ensuring a balance between predictive accuracy and consistency with biological principles. The framework is validated using the commercial UVA/Padova simulator, outperforming traditional linear models in glucose prediction accuracy and reconstruction of unmeasured states, even under circadian variations in insulin sensitivity. The results demonstrate the potential of BIRNN for personalized glucose regulation and future adaptive control strategies in AP systems.