Image Acquisition Context
Procedure Description Attributes for Clinically Relevant Indexing and Selective Retrieval of Biomedical Images
- W Dean Bidgood Jr,
- Bruce Bray,
- Nicolas Brown,
- Angelo Rossi Mori,
- Kent A Spackman,
- Alan Golichowski,
- Robert H Jones,
- Louis Korman,
- Brent Dove,
- Lloyd Hildebrand,
- Michael Berg
- Affiliations of the authors: University of Oklahoma, Oklahoma City, Oklahoma (WDB, LH); University of Utah, Salt Lake City, Utah (BB); University College London Medical School, London, England (NB); Istituto Tecnologie Biomediche, CNR, Rome, Italy (ARM); Oregon Health Sciences University, Portland, Oregon (KAS); Indiana University School of Medicine, Indianapolis, Indiana (AG); Duke University, Durham, North Carolina (RHJ); Washington Veterans Administration Medical Center, Washington, D.C. (LK); University of Texas Health Science Center, Austin, Texas (BD); University of Rochester, Rochester, New York (MB)
- Corresdpondence and reprints: W. Dean Bidgood, Jr, MD, MS, 5 Osborne Place, Durham, NC 27705. e-mail: 〈bidgood{at}nlm.nih.gov〉
- Received 15 December 1997
- Accepted 17 August 1998
Abstract
Objective To support clinically relevant indexing of biomedical images and image-related information based on the attributes of image acquisition procedures and the judgments (observations) expressed by observers in the process of image interpretation.
Design The authors introduce the notion of “image acquisition context,” the set of attributes that describe image acquisition procedures, and present a standards-based strategy for utilizing the attributes of image acquisition context as indexing and retrieval keys for digital image libraries.
Methods The authors' indexing strategy is based on an interdependent message/terminology architecture that combines the Digital Imaging and Communication in Medicine (DICOM) standard, the SNOMED (Systematized Nomenclature of Human and Veterinary Medicine) vocabulary, and the SNOMED DICOM microglossary. The SNOMED DICOM microglossary provides context-dependent mapping of terminology to DICOM data elements.
Results The capability of embedding standard coded descriptors in DICOM image headers and image-interpretation reports improves the potential for selective retrieval of image-related information. This favorably affects information management in digital libraries.
Footnotes
-
This work was supported in part by the National Library of Medicine, the Joseph F. Stein Foundation, the American College of Radiology, the American Academy of Ophthalmology, the American Dental Association, the American Society for Gastrointestinal Endoscopy, the American Academy of Neurology, and the College of American Pathologists.
-
↵* For a more detailed explanation of the notation, see Coad and Yourdon.19
