<ul><li>A study introduces a Biological-Informed Recurrent Neural Network (BIRNN) framework for accurate glucose-insulin dynamics modeling.</li><li>The BIRNN leverages a Gated Recurrent Units (GRU) architecture augmented with physics-informed loss functions.</li><li>The framework outperforms traditional linear models in glucose prediction accuracy and reconstruction of unmeasured states.</li><li>The results demonstrate the potential of BIRNN for personalized glucose regulation and adaptive control strategies in Artificial Pancreas (AP) systems.</li></ul>

Integrating Biological-Informed Recurrent Neural Networks for Glucose-Insulin Dynamics Modeling

Discover more