Integrating image data into biomedical text categorization

Abstract

Categorization of biomedical articles is a central task for supporting various curation efforts. It can also form the basis for effective biomedical text mining. Automatic text classification in the biomedical domain is thus an active research area. Contests organized by the KDD Cup (2002) and the TREC Genomics track (since 2003) defined several annotation tasks that involved document classification, and provided training and test data sets. So far, these efforts focused on analyzing only the text content of documents. However, as was noted in the KDD'02 text mining contest-where figure-captions proved to be an invaluable feature for identifying documents of interest-images often provide curators with critical information. We examine the possibility of using information derived directly from image data, and of integrating it with text-based classification, for biomedical document categorization. We present a method for obtaining features from images and for using them-both alone and in combination with text-to perform the triage task introduced in the TREC Genomics track 2004. The task was to determine which documents are relevant to a given annotation task performed by the Mouse Genome Database curators. We show preliminary results, demonstrating that the method has a strong potential to enhance and complement traditional text-based categorization methods.

Structured Literature Image Finder: Parsing Text and Figures in Biomedical Literature

“…As part of our work for the Elsevier Challenge, we expanded the classification to other panel types. This mirrors other systems that have appeared since the original SLIF which include more panel types [7,8,9].…”

Improving Product Classification Using Images

“…Although classification using generic multi-modal signals has been recently proposed in the literature, the use of images to aid text classification in a setting where there is significant difference in discriminative power between the two types of signals has not been systematically studied before. In fact, the only previous work [20] we are aware of is in the context of document classification. In that setting, several images in the document are used in conjunction with the text to identify the true category of a document.…”

Exploring the efficacy of caption search for bioscience journal search interfaces

“…Srihari (1991;1995) did early work on linking information between photographs and their captions, to determine, for example, which person's face in a newspaper photograph corresponded to which name in the caption. Shatkay et al (2006) (Cohen et al, 2003) they devised and tested algorithms for parsing the structure of image captions, which are often quite complex, especially when referring to a figure that has multiple images within it. In another effort, they developed tools to extract information relating to subcellular localization by automatically analyzing fluorescence microscope images of cells .…”