Determining the complexity of Dutch medical terms by using a familiarity metric
Keywords
Loading...
Authors
Issue Date
2022-06-19
Language
en
Document type
Journal Title
Journal ISSN
Volume Title
Publisher
Title
ISSN
Volume
Issue
Startpage
Endpage
DOI
Abstract
Patients are getting better access to their own medical records due to the rise of Electronic
Health Records (EHRs). However, these records often do not have the right readability
level for every patient to understand what is being said. To help increase the readability
of the EHRs, a familiarity metric was created to measure the complexity of Dutch medical
terms. This metric was developed using the frequencies found in different text corpora.
Afterwards, the familiarity metric was validated by use of a survey asking participants
to give the complexity of a sample of medical terms on a seven-point Likert scale. From
the results it followed that a positive correlation was found with a Kendall’s τc of 0.53
between the mean values given from the survey and the scores generated by the algorithm.
Therefore, it can be concluded that using a frequency-based familiarity metric is a good
way to measure the complexity of Dutch medical terms. In future research, the metric
can be expanded by taking into account multiple features, such as abbreviations and the
use of affixes. Besides, this study is a good step in the direction of improving an earlier
proposed algorithm that generates clarifications of diagnoses. This can help identify the
complex medical terms used in EHRs, such that these words can either be clarified or
replaced with easier synonyms in order to improve the readability of these records.
Description
Citation
Supervisor
Faculty
Faculteit der Sociale Wetenschappen