Insulin dosing of hyperglycemic patients in the intensive care unit (ICU) is a complex and nonlinear clinical control problem. Recent model-based glycemic control protocols predict a patient-specific and time-specific future insulin sensitivity distribution, which defines the future patient state in response to insulin and nutrition inputs. The prediction methods provide a 90% confidence interval for a future insulin sensitivity distribution for a given time horizon, making the prediction problem more specific compared to common prediction problems where the aim is to predict the expected value of the given stochastic parameter. This study proposes three alternative artificial intelligence-based insulin sensitivity prediction methods to improve the prediction accuracy and make prediction parameters better fit the clinical requirements. The proposed prediction methods use different neural network models: a classification deep neural network model, a Mixture Density Network model, and a Quantile Regression-based model. A large patient data set was used to create the neural network models, including 2357 patients and 92646 blood glucose measurements from three clinical sites (Christchurch, New Zealand, Gyula, Hungary, and Liege, Belgium). Prediction accuracy was assessed by statistical metrics expressing clinical requirements, as well as via validated in-silico virtual patient simulations comparing the clinical performance of a proven glycaemic control protocol using the alternative prediction methods to assess impact on glycemic control performance and thus the need for these alternative models.

High blood glucose levels and stress-induced hyperglycemia are common issues in intensive care units (ICU). Controlling blood glucose levels is crucial but challenging due to patient-specific variability. This challenge was addressed by developing model-based control protocols, which rely on identifying and predicting the patient-specific metabolic state. This study presents the in-silico simulation-based evaluation of a new artificial neural network-based insulin sensitivity (SI) prediction method. The models were trained on a dataset collected during clinical treatment using the stochastic-targeted (STAR) protocol and evaluated by simulating the clinical interventions on virtual patients created from retrospective clinical data. The results show the new models could be safely applied for SI prediction. Furthermore, the adopted method had very similar accuracy in the overall comparison of cohorts, with only minor differences noted in hypoglycemia events.

Current Blood Glucose (BG) monitoring techniques are invasive or semi-invasive and can impose financial and practical burden on patients. In this study a compact, non-invasive and single port Substrate Integrated Waveguide (SIW) loaded with X-shape has been present. The sensor resonant at 1.7 GHz, a dual parameter employed to evaluate the sensor by tracking the shift of resonance frequency in (MHz) and the reflection coefficient (dB) to detect glucose-induced changes in tissue permittivity. The device is designed using full-wave electromagnetic simulations with multilayer tissue models and validated experimentally on five human volunteers under controlled fasting and post-glucose conditions. Across the physiological range of 20–200 mg/dL, the sensor exhibits sensitivities up to 0.310 MHz per mg/dL and 0.333 dB per mg/dL, demonstrating consistent responsiveness to glucose variations. The results indicate that the proposed resonator can track glucose-related dielectric changes using a simple contact-based configuration. However, the measurements are influenced by subject-specific variability and sensor placement conditions, which currently limit generalization and repeatability. Further work is required to improve robustness, calibration, and validation on larger cohorts before practical deployment.