Open Self Medication on LOD

Open Self Medication is a Web application that aims to accompany the general public in his initiative to self-medicate, i.e. the act of treating undiagnosed medical ailments with unprescribed drugs. The application achieves this goal by providing a set of functionalities that ensure safety and efficiency of this practice. With safety, we mean that the system guides the end-user from a set of common mild medical signs to adapted molecules and drug products, but also highlights the risks, e.g. drug interactions, adverse events, of self prescribing a drug in a given situation. The efficiency argument corresponds to providing a rating, based on a tolerance/efficiency ratio designed by a team of health care professionals, to some identified self-medication molecules. A main characteristic of this application is that almost all the data processed by the system and presented to the end-user comes from a subset of the LOD datasets, namely DrugBank, DailyMed, Sider and DBPedia. This paper motivates the design of such an application, provides the main design choices, describes some implementation details and presents lessons learned and future work.