论文信息 - Indexing Large Archives of Pathology Images Using the Unified Medical Language System (UMLS)

Indexing Large Archives of Pathology Images Using the Unified Medical Language System (UMLS)

The value of any large image archive resides in the ability to select and retrieve images based on features of interest contained in the images. We here show that images can be automatically encoded from descriptive text (image-legends), into concept codes of the Unified Medical Language System (UMLS), a technique that permits powerful image categorization and retrieval, and is generalizable to image archives of any size. A collection of 5,465 pathology image legends was encoded into UMLS terms, via our computer translation program that parses and maps plain-text image-legends into lists of UMLS terms. Each image-legend yielded an average of 15 indexterms, ranging in frequency from five terms in the least-indexed legend to 58 terms in the most-indexed legend. The resulting UMLS index can be used to retrieve images, even when a chosen query term is not included in the image legend.

Jules J. Berman | G. William Moore | David S. Brenner

[1] Charles Safran,et al. Validation of clinical problems using a UMLS-based semantic parser , 1998, AMIA.

[2] G W Moore,et al. Automatic SNOMED coding. , 1994, Proceedings. Symposium on Computer Applications in Medical Care.

[3] Henry J. Lowe,et al. Towards knowledge-based retrieval of medical images. The role of semantic indexing, image content representation and knowledge-based retrieval , 1998, AMIA.

[4] G W Moore,et al. SNOMED-encoded surgical pathology databases: a tool for epidemiologic investigation. , 1996, Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc.

[5] Gunther Eysenbach,et al. An International Dermatological Image Atlas on the WWW: Practical Use for Undergraduate and Continuing Medical Education, Patient Education and Epidemiological Research , 1998, MedInfo.

[6] Patrice Degoulet,et al. A property concept frame representation for flexible image-content retrieval in histopathology databases , 2000, AMIA.

[7] G W Moore,et al. Performance analysis of manual and automated systemized nomenclature of medicine (SNOMED) coding. , 1994, American journal of clinical pathology.

[8] J. Berman,et al. A SNOMED analysis of three years' accessioned cases (40,124) of a surgical pathology department: implications for pathology-based demographic studies. , 1994, Proceedings. Symposium on Computer Applications in Medical Care.