A web-based medical text simplification tool

Abstract

With the increasing demand for improved health literacy, better tools are needed to produce personalized health information efficiently that is readable and understandable by the patient. In this paper, we introduce a web-based text simplification tool that helps content-producers simplify existing text materials to make them more broadly accessible. The tool uses features that provide concrete suggestions and all features have been shown individually to improve the understandability of text in previous research. We provide an overview of the tool along with a quantitative analysis of the impact on medical texts. On a medical corpus, the tool provides good coverage with suggestions on over a third of the words and over a third of the sentences. These suggestions are over 40% accurate, though the accuracy varies by text source.