Automated semantic indexing of imaging reports to support retrieval of medical images in the multimedia electronic medical record

This paper describes preliminary work evaluating automated semantic indexin g of radiology imaging reports to represent images stored in the Image Engi ne multimedia medical record system at the University of Pittsburgh Medical Center, The authors used the SAPHIRE indexing system to automatically iden tify important biomedical concepts within radiology reports and represent t hese concepts with terms from the 1998 edition of the U.S. National Library of Medicine's Unified Medical Language System (UMLS) Metathesaurus. This a utomated UMLS indexing was then compared with manual UMLS indexing of the s ame reports. Human indexing identified appropriate UMLS Metathesaurus descr iptors for 81% of the important biomedical concepts contained in the report set. SAPHIRE automatically identified UMLS Metathesaurus descriptors for 6 4% of the important biomedical concepts contained in the report set. The ov erall conclusions of this pilot study were that the UMLS metathesaurus prov ided adequate coverage of the majority of the important concepts contained within the radiology report test set and that SAPHIRE could automatically i dentify and translate almost two thirds of these concepts into appropriate UMLS descriptors. Further work is required to improve both the recall and p recision of this automated concept extraction process.