An ontology for breast radiologist training.
Presented at: UNSPECIFIED.
Computer-based tools are increasingly used in radiologists' training and numerous initiatives are attempting to compile large databases of annotated digital images for training purposes. We want to apply computational intelligence to support individualised learning in mammography, based on a set of annotated cases. We require not just a controlled set of radiological descriptors to be used in the annotation but rather an ontology to support reasoning about the educational content of a case. Previous research in biomedical ontologies has modelled normal anatomy, genetic processes and other clinical concepts. There has, however, been relatively little work on imaging or on modelling educational concepts. In this paper, we describe an ontology for breast radiologist training. The ontology is based on a thorough study of 400 cases, selected and annotated by experienced screening radiologists. In addition to the usual reporting information, annotations also include learning points, indicating their educational significance. We give a detailed description of our ontology, and of how the learning points are interpreted and represented. The novel ontology enables computerbased reasoning, allowing a teaching tool to deliver an individualised training programme.
|Type:||Conference item (UNSPECIFIED)|
|Title:||An ontology for breast radiologist training|
|UCL classification:||UCL > School of Life and Medical Sciences
UCL > School of Life and Medical Sciences > Faculty of Population Health Sciences
UCL > School of Life and Medical Sciences > Faculty of Population Health Sciences > Institute of Epidemiology and Health Care > CHIME
Archive Staff Only