A Language Independent Decision Support System for Diagnosis and Treatment by Using Natural Language Processing Techniques

摘要

Global mobility including all countries in the world, is growing at a rate faster than the world's population, which is surpassed 244 million people in 2015. The explosive growth in the importance, interest and the study of international migration urges new developments for accessible, efficient and affordable global public health systems. In order to actualize such improvements and eliminate the undesirable consequences of wrong or late diagnosis in global public healthcare, we propose an intelligent healthcare diagnostics engine. Current diagnosis systems have obstacles about easy user access by worldwide heath seekers. In this study, it is aimed to reach patients using different languages while providing an opportunity to enter symptoms in their everyday language text, besides medical expressions of symptoms. Language independency is provided on the background of user interface by using translate functions in TextBlob, python. Named entity recognition (NER) techniques, based on natural language processing (NLP), are applied to develop language independent predictive model. Extracted terms are implied as an input of the model and analyzed for degree of matching symptoms for the corresponding diagnosis. The accuracy at a range of 20 and 100% has been accomplished based on the degree of matching of patient's enquiry with database of the system, for languages other than English.