RSSCEB Projects |
 Print this
 E-mail this
|
Image & Text IntegrationProject Member(s): Sameer Antani, Dina Demner-Fushman, Matthew Simpson, George Thoma
Medical images are essential in establishing diagnosis, analyzing and evaluating treatment results, and as educational materials in many clinical specialties. Despite the fact that in such specialties as dermatology, trauma surgery, and radiology images are routinely generated for the above purposes and subsequently used in publications, to date, not many studies are dedicated to providing images for clinical decision support.
The goal of the Image and Text Integration (ITI) project is to automatically annotate biomedical images extracted from scientific publications with respect to their modality, content, usefulness for clinical decision support and instructional purposes, and project the annotations onto unannotated images stored in databases or obtained in a clinical setting by linking images through content-based image retrieval (CBIR). This goal is attainable through a judicious combination of text-based decision support RIDeM, InfoBot and semantic annotation and retrieval of EBP-relevant images from scientific publications. Automatic image annotation and subsequent retrieval can be based purely on image analysis, on indexing of the text assigned to images, or on a combination of image and text analysis. More...
National Institutes of Health (NIH)
U.S. Dept. of Health