Web based prediction for diabetes treatment

Diabetics need continuous support during their therapy because the clinical treatment could be improved in the medical cures;the doctors need to share the new information and to support their patients in an interactive way. In this paper, we describe a web based software tool able to perform predictions on the future glycaemia level of the patients for a specific time horizon; to perform what-if analysis, showing how the glycaemia level could vary if some parameters are changed (either in the diet of the patient or in the clinical treatment); to share information and suggestions, creating a virtual community between specialists and patients; to inform users about their health condition, by a classification algorithm and to allow the patient to maintain a daily diary and to know about the evolution of the health condition. The system merges software engineering and Operation Research methods and could become a valid interactive support for patients and specialists.