Tratamento semântico de registros eletrônicos sobre cuidados de AVC para um modelo de gestão baseado em valor

Abstract

Value-based health managment models require a precise accounting of health indexes such as risk events monitoring, clinical conditions, pacient handling and, cases disclosures. Currently this accounting is performed by manually reading and searching through eletronic health records for this indexes. Our research proposes a way to make this a autonomous task that can be performed by a computer using a free-text concept classifier model based on ontologies. To validate our model we tested it with digital clinical evaluations from 281 pacients under stroke care. We’ve selected 30 indexes to be identified in this texts. Our model was capable of identify and classify 28 of thoses indexes varying from ’5,83 % f1-score results and mcc score of 8,01 % to ’94,78 % f1-score results and mcc-score of 94,78 %’. Considering all 30 indexes, our model reached, on average ’56,8 % of f1-score and a mccscore of 57,97 %’.