Abstract:
Diabetes is among one of the fastest growing diseases all over the world. Controlled diet and regular exercise are considered as a treatment to control type 2 diabetes. However, food and exercise suggestions in existing solutions do not consider integrated knowledge from personal profile, preferences, current vital signs, diabetes domain, food domain and exercise domain. Moreover, existing conventional methods advices general recommendations those are not applicable to all variety of diabetes patients. Furthermore, the strong correlation between diet and exercise is ignored in current existing solutions. A single diet management scheme such as food pyramid, diet chart, carbohydrate counting and glycemic index doesn’t gives customized, personalized and balanced diet. Our approach uses combination of these diet management techniques. We have implemented an ontology based integrated approach to combine knowledge from various domains to generate diet and exercise suggestions for diabetics. The solution is developed as a Semantic Healthcare Assistant for Diet and Exercise (SHADE). For each domain (person, diabetes, food and exercise) we have defined separate OWL (Web Ontology Language) based ontology along with SWRL (Semantic Web Rule Language) rules and then an integrated ontology combines these individual ontologies. Our prototype application is developed using Java, Jena and Pellet API. Based on data and rules, Pellet reasoning engine semantically recommends diet and exercise recommendations as inferences. The research work presents diet recommendations in the form of various alternative customized menus such that each menu is a personalized, healthy and balanced diet. Finally, exercise recommendations are generated in the form of various alternative and user’s preferred physical activities along with duration and intensity.
