Textual Intelligence

19Toward Accurate Abbreviation Disambiguation in Medical Texts: A Comparative Study of AI Models

A. Pandey¹ and M. Saini²*

¹Institute of Applied Sciences and Humanities, GLA University, Mathura, Uttar Pradesh, India

²School of Engineering and Technology, BML Munjal University, Gurugram, Haryana, India

Abstract

Abbreviations are important for concise communication in the era of medical documentation because they enable healthcare providers to comply with patient privacy laws while conveying complex medical concepts. However, these abbreviations present a significant issue as they are inherently ambiguous, which means that their meaning can vary depending on the context in which they are used. This ambiguity leads to an increased chance of misinterpretation, which can ultimately harm patient care if the incorrect expansion is applied. It is challenging to extend medical abbreviations exactly because there is no common mapping technique. To consistently identify the correct expansions of medical abbreviations and minimize these barriers while ensuring the appropriate interpretation of these abbreviations, there is a need for an intelligent system that can successfully learn from a variety of textual material, including words, phrases, acronyms, and abbreviations. The best performing model for this problem is determined by carefully comparing the significance of various state-of-the-art models, including support vector machine, random forest, K-nearest neighbor, decision tree, ...