FactCheXcker: Mitigating Measurement Hallucinations in Chest X-ray Report Generation Models

TL;DR

FactCheXcker is a modular framework that reduces measurement hallucinations in chest X-ray report generation by leveraging large language models and specialized modules, significantly improving measurement accuracy across multiple models.

Contribution

Introduces FactCheXcker, a novel framework that de-hallucinates radiology report measurements using an improved query-code-update paradigm with large language models.

Findings

Significantly reduces measurement hallucinations in models.

Improves measurement precision by an average of 135%.

Enhances 10 out of 11 evaluated report-generation models.

Abstract

Medical vision-language models often struggle with generating accurate quantitative measurements in radiology reports, leading to hallucinations that undermine clinical reliability. We introduce FactCheXcker, a modular framework that de-hallucinates radiology report measurements by leveraging an improved query-code-update paradigm. Specifically, FactCheXcker employs specialized modules and the code generation capabilities of large language models to solve measurement queries generated based on the original report. After extracting measurable findings, the results are incorporated into an updated report. We evaluate FactCheXcker on endotracheal tube placement, which accounts for an average of 78% of report measurements, using the MIMIC-CXR dataset and 11 medical report-generation models. Our results show that FactCheXcker significantly reduces hallucinations, improves measurement precision, and maintains the quality of the original reports. Specifically, FactCheXcker improves the performance of 10/11 models and achieves an average improvement of 135.0% in reducing measurement hallucinations measured by mean absolute error. Code is available at https://github.com/rajpurkarlab/FactCheXcker.