TIDA: A spanish EHR semantic search engine

Abstract

Electronic Health Records (EHR) and the constant adoption of Information Technologies in healthcare have dramatically increased the amount of unstructured data stored. The extraction of key information from this data will bring better caregivers decisions and an improvement in patients’ treatments. With more than 495 million people talking Spanish, the need to adapt algorithms and technologies used in EHR knowledge extraction in English speaking countries, leads to the development of different frameworks. Thus, we present TIDA, a Spanish EHR semantic search engine, to give support to Spanish speaking medical centers and hospitals to convert pure raw data into information understandable for cognitive systems. This paper presents the results of TIDA’s Spanish EHR free-text treatment component with the adaptation of negation and context detection algorithms applied in a semantic search engine with a database with more than 30,000 clinical notes.