Towards a Simplification of Medical Documents

Abstract

In this paper, we propose a solution to simplify reading and understanding medical documents via the automatic demystification of complex medical terms found in web pages. The suggested approach detects those terms using a combination of NLP and heuristics. It computes the probability that a word is a medical and complex term through text processing and analysis. It then makes use of an ad hoc dictionary to simplify the complex terms. Finally, it exploits the Microsoft Bing cognitive API to retrieve images and videos related to the complex medical terms detected and presents them to the users so that they may have a better understanding of the document being read.