Navigation in medical Internet image databases.

Area of Science:

• Medical Informatics
• Information Science

Background:

• The World Wide Web (WWW) revolutionizes database access, enabling integrated information presentation.
• The WWW is conceptualized as a database, necessitating analysis of navigation and query techniques.
• Medical image repositories on the WWW present diverse navigation methods and limited standardized querying.

Purpose of the Study:

• To analyze navigation techniques and semantic query types in medical image repositories on the WWW.
• To evaluate the potential of standardized vocabularies and semantic networks for improving WWW database navigation.

Main Methods:

• Searched for 48 medical image repositories using search engines.
• Analyzed navigation techniques (HTML link-lists, applets) and meta-information encoding (ACR, UMLS).
• Assessed the utility of standardized vocabularies (ICD10, Terminologia Anatomica) and the Unified Medical Language System (UMLS).

Main Results:

• Navigation methods varied widely, with applets showing promise for improved user experience.
• Query meta-information primarily used ACR and UMLS encoding; standardized vocabularies were largely absent.
• The UMLS Metathesaurus aids navigation, but its full potential is unrealized without integrating the UMLS Semantic Network.

Conclusions:

• Current medical image repositories on the WWW offer inconsistent navigation and querying capabilities.
• Implementing robust semantic networks, such as the UMLS Semantic Network, can significantly enhance navigation and access to medical multimedia resources.
• Standardized vocabularies and semantic relationships are crucial for developing more effective WWW-based medical databases.