Natural language Processing and Ontology based Decision Support System for Diabetic Patients

Diagnosis of diabetes disease is very promising because it may create various other acute or chronic health problems in the human body. This study proposes a recommendation system for diagnosis and treatment of Diabetes Mellitus as suggested by the Clinical Practice Guidelines (CPG). A treatment plan is designed which includes medications and lifestyle changes, according to the patient information, symptoms, and lab tests. Various tasks such as gathering, modelling, and implementing diabetic domain expertise knowledge collected from the experts, CPG, and other sources are used to construct an ontology for diabetes and a decision support system to manage diabetes in its early stages. This work also uses an Optical Character Recognition (OCR) and a Natural Language Processing (NLP) model that help directly to read doctor's prescription and extract useful information useful for developing the ontology. Web Ontology Language (OWL) is used to build the proposed ontology, and the Semantic Web Rule Language (SWRL) is used to create the rules for providing restrictions on the ontology.