IRMA Code II
Content-based image retrieval (CBIR) provides novel options to access large repositories of medical images. Thus, there are new opportunities for storing, querying and reporting especially within the field of digital radiology. This, however, requires a revisit of nomenclatures for image classification. The Digital Imaging and Communication in Medicine (DICOM), for instance, defines only about 20, partly overlapping terms for coding the body region. In 2002, the Image Retrieval in Medical Applications (IRMA) project has proposed a mono-hierarchic, multi-axial coding scheme. Although the initial concept of the IRMA Code was designed for later expansion, the appliance of the terminology in the practice of scientific projects discovered several weak points. In this paper, based on a systematic analysis and the comparison with other relevant medical ontologies such as the Lexicon for Uniform Indexing and Retrieval of Radiology Information Resources (RadLex), we accordingly propose axes for medical equipment, findings and body positioning as well as additional flags for age, body part, ethnicity, gender, image quality and scanned film. The IRMA Code II may be used in the Cross Language Evaluation Campaign (CLEF) annotation tasks as a database of classified images to evaluate visual information retrieval systems.
Unable to display preview. Download preview PDF.
- 5.Zaiß A, Graubner B, Ingenerf J, et al. Medizinische Dokumentation, Terminologie und Linguistik. In: Lehmann TM, editor. Handbuch der Medizinischen Informatik. 1st ed. Aachen: Hanser; 2002. p. 45–99.Google Scholar
- 6.Thun S. Medizinische Dokumentation und Kommunikation. In: Johner C, Haas P, editors. Praxishandbuch IT im Gesundheitswesen: Erfolgreich einf¨uhren, entwickeln, anwenden und betreiben. 1st ed. M¨unchen: Hanser; 2009. p. 131–60.Google Scholar
- 7.Mildenberger P, Eichelberg M, Martin E. Introduction to the DICOM standard. Eur Radiol. 2002 Apr;12(4):920–7.Google Scholar