Abstract
In biomedical publications, figures and images concisely summarize a paper's experimental findings and results. Recent studies have therefore explored the use of images to assist in information retrieval (IR) in biomedicine, mostly based on mining the image caption content. We extend this approach by mining the image text, which refers to the text inside biomedical figures and images. In this work, we discuss the distinct advantages of using image text for biomedical IR and present a prototype search engine implementing the idea.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 1186 |
| Number of pages | 1 |
| Journal | AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium |
| State | Published - 2008 |
| Externally published | Yes |
All Science Journal Classification (ASJC) codes
- General Medicine