Related Experiment Video
Updated: Jun 4, 2026

Cloud-Based Phrase Mining and Analysis of User-Defined Phrase-Category Association in Biomedical Publications
Published on: February 23, 2019
Evaluating the Importance of Image-related Text for Ad-hoc and Case-based Biomedical Article Retrieval
Matthew S Simpson1, Dina Demner-Fushman, George R Thoma
1Lister Hill National Center for Biomedical Communications, U.S. National Library of Medicine, Bethesda, MD 20894.
Abstract:
Images and their associated text are an essential source of information in biomedical articles. However, their use in providing evidence for clinical case descriptions has yet to be evaluated in the context of information retrieval. Given the complexity of case-based document retrieval, understanding the importance of images and image-related text is critical for future research into text- and content-based approaches to this problem. In this study, we compare the extent to which image-related text is useful in facilitating document retrieval for both case-based information requests and ad-hoc clinical questions in the domain of family practice. We show that case-based document retrieval is significantly improved with the use of image-related text whereas retrieval for clinical questions is largely unaffected. This suggests that visual evidence is more relevant for the case descriptions used in our study than the clinical questions.

