Evidence-Grounded Chest X-ray Report Generation with Retrieval, Citation, and Hallucination Control
Jurnal Sains dan Kesehatan · 31.12.2023 · 7 (2) · 53-74
DOI: 10.57214/jusika.v7i2.1127 ↗Kurz gefasst
Fachartikel in der Zeitschrift Jurnal Sains dan Kesehatan, online erschienen am 31.12.2023. Verfasst von Danang Danang und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Chest X-ray report generation has become an important topic in vision-to-language research. However, fully generative models often create fluent clinical reports that may contain unsupported or inaccurate statements, leading to hallucination problems and reducing reliability. This study investigates evidence-grounded report generation using the Open-i (IU X-ray) dataset with two main goals: generating coherent radiology reports from X-ray images and minimizing unsupported clinical entities through evidence retrieval. Four experimental models were evaluated: a baseline image-to-report model (E1), an alignment-enhanced model using the InfoNCE objective (E2), a retrieval-grounded model that incorporates Top-K evidence sentences with citation markers such as [E#] (E3), and a reranking model that selects the most evidence-supported output (E4). Experimental results on the Open-i test dataset show that grounding methods significantly reduce hallucination rates and improve entity overlap performance. The reranking approach achieves the best grounding quality, although stronger grounding slightly lowers text-overlap scores and increases inference time. Overall, retrieval-based grounding with explicit citations and reranking offers an effective approach for improving factual consistency and reducing unsupported information in automated radiology report generation.
Zitieren
APA
Danang, D., & Putra, T. W. A. (2023). Evidence-Grounded Chest X-ray Report Generation with Retrieval, Citation, and Hallucination Control. Jurnal Sains dan Kesehatan, 7(2), 53-74. https://doi.org/10.57214/jusika.v7i2.1127
Vancouver
Danang D, Putra TWA. Evidence-Grounded Chest X-ray Report Generation with Retrieval, Citation, and Hallucination Control. Jurnal Sains dan Kesehatan. 2023;7(2):53-74. doi:10.57214/jusika.v7i2.1127
BibTeX
@article{Danang2023,
author = {Danang, Danang and Putra, Toni Wijanarko Adi},
title = {Evidence-Grounded Chest X-ray Report Generation with Retrieval, Citation, and Hallucination Control},
journal = {Jurnal Sains dan Kesehatan},
year = {2023},
volume = {7},
number = {2},
pages = {53--74},
doi = {10.57214/jusika.v7i2.1127}
}
EUVIMED