A proposal for a LOINC automatic mapping support tool

Clinical coding and classification systems are the lingua franca to overcome idiosyncrasiesof the local information systems. Among those required into the Italian EHR, LOINCaims to uniquely identify both clinical document types indexed into EHR registries and testsinto laboratory reports. Its use requires that local test codes are mapped to the codes of thestandard so that equivalent concepts are aligned and can be easily understood and reused byother systems. This mapping process is cost and time-consuming as involves for many hoursdomain experts to find the right association. Therefore, this paper aims to make a proposalfor an automatic mapping tool to support domain experts in finding the correct LOINC codeto map to, specifically customized on the Italian language, combining textual similarity techniqueswith new NLP models and taking advantage of the so-called "wisdom of the crowd".