New search engine optimized for electronic teaching files debuts

December 1, 2006

A prototype search engine uses advanced content-specific algorithms to efficiently search databases of image files contained in public Medical Imaging Resource Center storage servers. The search engine was developed as a joint project between the Baltimore VA Medical Center and the University of Maryland. The prototype is the alpha version of a vertical search engine designed to match users with radiology-specific content.

A prototype search engine uses advanced content-specific algorithms to efficiently search databases of image files contained in public Medical Imaging Resource Center storage servers. The search engine was developed as a joint project between the Baltimore VA Medical Center and the University of Maryland. The prototype is the alpha version of a vertical search engine designed to match users with radiology-specific content.

Popular search engines cannot provide comprehensive compilations of teaching files in MIRC sites or other radiology information sites that contain case examples. Teaching files are "hidden" from a typical search engine because they are located behind pages of fill-in-the-blank forms.

GoogleMIRC has the ability to obtain and index online storage teaching files located behind these forms. This Web crawler also protects itself from dynamic content, advertisements, and unwanted downloads. Its crawler speed is 1.5 pages per second.

GoogleMIRC provides a portal for searching for specific pages, using a complex algorithm to determine whether each case is relevant. The search engine currently uses 13 servers. In demonstrations at the Lakeside Learning Center, it outperformed commercial search engines by a significant margin.

Related Content:

RSNA