by Cindy Li, Elizabeth Chen, Guergana Savova, Hamish Fraser and Carsten Eickhoff
Diagnostic errors can pose a serious threat to patient safety, leading to serious harm and even death. Efforts are being made to develop interventions that allow physicians to reassess for errors and improve diagnostic accuracy. Our study presents an exploration of misdiagnosis patterns mined from PubMed abstracts. Article titles containing certain phrases indicating misdiagnosis were selected and frequencies of these misdiagnoses calculated. We present the resulting patterns in the form of a directed graph with frequency-weighted misdiagnosis edges connecting diagnosis vertices. We find that the most commonly misdiagnosed diseases were often misdiagnosed as many different diseases, with each misdiagnosis having a relatively low frequency, rather than as a single disease with greater probability. Additionally, while a misdiagnosis relationship may generally exist, the relationship was often found to be one-sided.
This repository contains the complete codebase producing all results of our study.
This paper appeared at the AMIA Informatics Summit 2020 and can be found here: https://arxiv.org/abs/2006.13721
If you would like to cite this work, please refer to:
@INPROCEEDINGS{li2020mining,
title={{Mining Misdiagnosis Patterns from Biomedical Literature}},
author={Li, Cindy and Chen, Elizabeth and Savova, Guergana and Fraser, Hamish and Eickhoff, Carsten},
booktitle={{Proceedings of the AMIA Informatics Summit}},
year={2020},
organization={AMIA}
}