This work introduces a safe data-driven control methodology, Data-Enabled Predictive Control (DeePC), for the control of blood glucose in type-1 diabetic patients. DeePC utilizes input-output trajectory data directly without requiring a system model or state estimation like other modelbased algorithms. The control strategy is validated using the Bergman Minimal Model, a well-established framework for glucose-insulin dynamics. Comparative simulations are conducted against Proportional-Integral-Derivative (PID) and Model Predictive Control (MPC) strategies. Results show that DeePC achieves comparable or superior glycemic regulation, particularly under model uncertainty, by maintaining normoglycemia and reducing hypoglycemia risk. The findings demonstrate the robustness and potential of DeePC in biomedical applications where model accuracy is uncertain. Future works would include computational efficiency improvements and handling uncertainties in meal estimation.

Type 1 diabetes is a growing global health challenge. Standard clinical practice often relies on manual insulin injections, which can lead to suboptimal glucose regulation. Recent advancements have shifted focus towards Artificial Pancreas systems, integrating continuous glucose monitoring with automated insulin delivery. This work presents a quantitative comparison of four Deep Reinforcement Learning algorithms for autonomous glycemic regulation via insulin injection: DDPG, PPO, SAC, and TD3. The validation is conducted using the Hovorka model, in presence of uncertainties on number, time and amount of meals. Results show that all four controllers are able to maintain blood glucose levels within the target range. The TD3 algorithm outperforms the others in terms of several key performance indicators such as time in range, time in hypo/hyperglycemia and total insulin usage, while also exhibiting fewer hyperglycemic episodes compared to prior works in academic literature.

In this paper, we investigate the use of Physics-Informed Neural Networks (PINNs) for the solution of an optimal control problem related with the optimal control of a stationary electric storage system (ESS) installed in a service station for plug-in electric vehicles (PEVs). The ESS is used to balance the PEVs charging power, in order to mitigate the impact on the grid, and keep low the power flow at the point of connection with the grid. The proposed PINN is trained in order to learn the optimality conditions of the optimal control problem so that, after training, it can provide the solution with no significant computation effort. This one represents a promising alternative to the analytical computation of the optimal control (which is possible only in very simple settings), and to the solution of the optimal control problem with numerical methods, which requires significant time in the more complex and realistic settings. Numerical simulations are presented to evaluate the effectiveness of the trained PINN in solving the optimal control problem.

Artificial Intelligence and Machine Learning have brought transformative changes to clinical diagnostics, especially in image classification via deep convolutional neural networks. The latter are crucial in analyzing and identifying diseases from medical visuals, ensuring accurate diagnosis and prevention. However, the limited availability of medical imaging data poses a serious challenge and leads to poor performance of the classifier. To address this issue, both geometric/color augmentation and synthetic data generation (through Generative Adversarial Networks and Variational Autoencoders) methods are employed, both operating on the entire dataset without an intrinsic optimization of the classification process. This study introduces an automated data augmentation strategy in which the editing operation is customized with respect to the individual images contained in the training set. This is done by using the Deep Reinforcement Learning framework provided by the Proximal Policy Optimization algorithm, with the reward being the test accuracy of the image classifier. The application of the proposed procedure on a meager dataset related to gastrointestinal diseases demonstrates an improvement in image classification by over 3%.