ABSTRACT
Taking care of diabetes means keeping track of symptoms and making changes based on what each person needs. This study proposes a novel approach to diabetes treatment. The study employs numerous digital twins and personal health knowledge graphs (PHKGs). We utilized PHKGs to construct a real-time digital twin system that prioritizes the patient's needs. We use HL7 standards to ensure the integration of information from various sources in this system. It makes it simple to access and share information, and it makes sure that the data and health findings are very accurate. PHKGs provide a file that is compatible with a wide range of tools. It is simple to add more information to the PHKG if more information about the patient comes to light. This enhances the precision and accuracy of the provided care. This method is flexible, which makes it easier to make new apps and supports continuous improvement. We used our digital twins in different diabetes control scenarios to demonstrate that they are flexible and work. Some of these include predicting glucose levels, finding the best insulin dose, making specific living suggestions, and visualizing health data. This study opens the door to more exact and personalized healthcare treatments by allowing real-time care tailored to each patient. This could lead to better long-term results for diabetes control.
