Cyber-Physical Systems (CPS) are crucial for developing Digital Twins (DT) due to their ability to bridge physical systems and virtual components. CPS integrate computational and physical processes with continuous feedback, traditionally analyzed through costly facilities. High-fidelity virtual models now serve as the basis for DT, which is emerging as a pivotal technology to realize CPS functions, such as to create bidirectional interactions between the physical and cyber realms. DT is driving significant progress on healthcare, where the abundance of data collected from numerous IoT devices and AI models enables: (i) the monitoring of patients’ lifestyles and regular eating habits and, (ii) providing alerts to patients about prescriptions and medical consultations, and (iii) assisting in selecting appropriate medications and predict the outcomes of surgical procedures. The healthcare sector increasingly demands advanced techniques to offer patients optimal assistance and drive digital transformation initiatives forward. This study seeks to elucidate and organize these guidelines, specifically aiming to clarify the integration of DT in healthcare. Finally, starting from an existing solution in literature, we propose an innovative preliminary medical model that relies on Geometric Deep Learning for the patient representation (the Patient DT) and Ontologies to facilitate the translation of complex healthcare data into a graph-based structure used by the Patient DT.
Healthcare and Cyberspace: from Cyber-Physical Systems to Medical Digital Twins
Rocco Zaccagnino;Delfina Malandrino;Gianluca Zaccagnino;Gerardo Benevento;
2025
Abstract
Cyber-Physical Systems (CPS) are crucial for developing Digital Twins (DT) due to their ability to bridge physical systems and virtual components. CPS integrate computational and physical processes with continuous feedback, traditionally analyzed through costly facilities. High-fidelity virtual models now serve as the basis for DT, which is emerging as a pivotal technology to realize CPS functions, such as to create bidirectional interactions between the physical and cyber realms. DT is driving significant progress on healthcare, where the abundance of data collected from numerous IoT devices and AI models enables: (i) the monitoring of patients’ lifestyles and regular eating habits and, (ii) providing alerts to patients about prescriptions and medical consultations, and (iii) assisting in selecting appropriate medications and predict the outcomes of surgical procedures. The healthcare sector increasingly demands advanced techniques to offer patients optimal assistance and drive digital transformation initiatives forward. This study seeks to elucidate and organize these guidelines, specifically aiming to clarify the integration of DT in healthcare. Finally, starting from an existing solution in literature, we propose an innovative preliminary medical model that relies on Geometric Deep Learning for the patient representation (the Patient DT) and Ontologies to facilitate the translation of complex healthcare data into a graph-based structure used by the Patient DT.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
