RSNA 2007 

Abstract Archives of the RSNA, 2007


LL-IN5206-B

GoldMiner: A Radiology Image Search Engine

Education Exhibits

Presented on November 25, 2007

Participants

Charles Edward Kahn MD, Presenter: Nothing to Disclose
Cheng Thao, Abstract Co-Author: Nothing to Disclose

PURPOSE/AIM

The purpose of this exhibit is to: Describe how structured concepts can be extracted from free-text figure captions Learn techniques to retrieve radiology images using concept- and keyword-based search techniques Identify educational and clinical applications of an Internet-based image library

CONTENT ORGANIZATION

Image search techniques Keywords Concepts Concept-based retrieval Organization of concepts: the UMLS Metathesaurus Discovering concepts in free-text figure captions Search engine Filtering search results by age, sex, imaging modality Applications Education Clinical decision support

SUMMARY

The major teaching points of this exhibit are: Automated tools can identify concepts in unstructured text of figure captions A search engine can combine keyword- and concept-based techniques to retrieve relevant images An Internet-based image library can enhance education and decision support in radiology

Cite This Abstract

Kahn, C, Thao, C, GoldMiner: A Radiology Image Search Engine.  Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL. http://archive.rsna.org/2007/5003994.html