Pubblicazioni

The integration of quantum computing into predictive modeling represents a transformative advancement in machine learning. This study compares classical Echo State Networks (ESN) with quantum Echo State Networks (qESN) for time-series forecasting, emphasizing the concept of a metabolic avatar - a dynamic data-driven model of individual metabolic processes. QESNs achieve comparable results to classical models under optimal conditions using a minimal number of qubits (4), outperform classical ESN significantly when trained with limited data, and demonstrate higher resilience to outliers.

This research presents the evolution from the Personalized Metabolic Avatar (PMA) to the Generalized Metabolic Avatar (GMA), advancing personalized weight forecasting through improved machine learning architectures and broader applicability across diverse patient populations.

A comprehensive study exploring the application of machine learning techniques in oral pathology diagnosis, demonstrating improved accuracy and efficiency in identifying pathological conditions.

This review examines digital applications for diet monitoring and precision nutrition, covering wearable devices, masticatory analysis, glucose monitoring systems, and mobile applications for dietary planning. The paper discusses the current state of technology and its potential for personalized nutrition interventions.

This study investigates the relationship between chewing patterns and glucose homeostasis using linear regression modeling, revealing significant correlations between mastication behaviors and postprandial glucose responses.

Development and evaluation of a non-invasive electromyographic device for analyzing personalized chewing patterns. The device evaluates chewing time, cycle time, work rate, number of chews and work, providing insights into mastication behaviors and their potential health implications.