ReXErr: Synthesizing Clinically Meaningful Errors in Diagnostic Radiology Reports
This addresses the challenge of improving radiology report quality for healthcare providers and AI developers, though it appears incremental as it builds on existing LLM capabilities for a specific domain application.
The researchers tackled the problem of errors in diagnostic radiology reports by introducing ReXErr, a methodology that uses Large Language Models to generate clinically plausible errors in chest X-ray reports, which demonstrated consistency across error categories and closely mimicked real-world errors.
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.