ReXErr: Synthesizing Clinically Meaningful Errors in Diagnostic Radiology Reports

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rao, Vishwanatha M., Zhang, Serena, Acosta, Julian N., Adithan, Subathra, Rajpurkar, Pranav
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929501496147968
author Rao, Vishwanatha M.
Zhang, Serena
Acosta, Julian N.
Adithan, Subathra
Rajpurkar, Pranav
author_facet Rao, Vishwanatha M.
Zhang, Serena
Acosta, Julian N.
Adithan, Subathra
Rajpurkar, Pranav
contents Accurately interpreting medical images and writing radiology reports is a critical but challenging task in healthcare. Both human-written and AI-generated reports can contain errors, ranging from clinical inaccuracies to linguistic mistakes. To address this, we introduce ReXErr, a methodology that leverages Large Language Models to generate representative errors within chest X-ray reports. Working with board-certified radiologists, we developed error categories that capture common mistakes in both human and AI-generated reports. Our approach uses a novel sampling scheme to inject diverse errors while maintaining clinical plausibility. ReXErr demonstrates consistency across error categories and produces errors that closely mimic those found in real-world scenarios. This method has the potential to aid in the development and evaluation of report correction algorithms, potentially enhancing the quality and reliability of radiology reporting.
format Preprint
id arxiv_https___arxiv_org_abs_2409_10829
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ReXErr: Synthesizing Clinically Meaningful Errors in Diagnostic Radiology Reports
Rao, Vishwanatha M.
Zhang, Serena
Acosta, Julian N.
Adithan, Subathra
Rajpurkar, Pranav
Computation and Language
Accurately interpreting medical images and writing radiology reports is a critical but challenging task in healthcare. Both human-written and AI-generated reports can contain errors, ranging from clinical inaccuracies to linguistic mistakes. To address this, we introduce ReXErr, a methodology that leverages Large Language Models to generate representative errors within chest X-ray reports. Working with board-certified radiologists, we developed error categories that capture common mistakes in both human and AI-generated reports. Our approach uses a novel sampling scheme to inject diverse errors while maintaining clinical plausibility. ReXErr demonstrates consistency across error categories and produces errors that closely mimic those found in real-world scenarios. This method has the potential to aid in the development and evaluation of report correction algorithms, potentially enhancing the quality and reliability of radiology reporting.
title ReXErr: Synthesizing Clinically Meaningful Errors in Diagnostic Radiology Reports
topic Computation and Language
url https://arxiv.org/abs/2409.10829