Semantics in radiology

This poster is published under an open license. Please read the disclaimer for further details.

Congress:

ECR 2015

Poster Number:

C-1335

Type:

Educational Exhibit

Keywords:

Computer applications, Image manipulation / Reconstruction, CAD, Computer Applications-Detection, diagnosis, Statistics, Education and training, Epidemiology

Authors:

H. Müller¹, E. Kotter²; ¹Sierre/CH, ²Freiburg/DE

DOI:

10.1594/ecr2015/C-1335

DOI-Link:

https://dx.doi.org/10.1594/ecr2015/C-1335

Fig. 1: Overview of semantic links between entities automatically extracted form a...

Fig. 2: Information retrieval interface that includes semantic data analysis as...

Fig. 3: Example of the Goldminer interface that uses RadLex for querying.

Learning objectives

understand the meaning of semantics in radiology understand the increasing importance that semantics can have in radiology understand current development in semantics in the radiology domain

Background

Definition: Generally, semantics is he study of meaning. Related to radiology, semantics denotes the study of the content or meaning of radiology reports, or – in a larger sense – of the content and meaning of images used in radiology. An image is worth a thousand words is often said and even more true is the phrase that an image can be described in a thousand different ways. Even in radiology where many things including the image creation are fairly standardized there are many ways...

Findings and procedure details

In radiology, RadLex [2] has been created to standardize annotation of images and thus reuse. One example of RadLex use is structured reporting where templates for many domains and types of observations exist and reusing the templates can significantly increase efficiency. Another area of RadLex use is information retrieval, where semantic knowledge of cases can make it much easier to find a specific case with a particular visual observation or a specific disease, again taking into account synonyms and also sub categories of a disease...

Conclusion

Semantics has started to become usable in many domains such as for the semantic web to allow machines to automatically process and understand digital content. In Radiology a semantic terminology exists, created mainly for writing radiology reports but also useful in making several parts of the radiologic workflow more effective and more efficient, such as semi-automatically preparing structured reporting templates and allowing access to relevant cases or to cases similar to a currently observed abnormality.

Personal information

Elmar Kotter, MD, PhD Abteilung Röntgendiagnostik, Universitätsklinikum Freiburg Hugstetterstr. 55, D-79106 Freiburg, Germany [email protected] Henning Müller, PhD, Prof. Information Systems Insitute, HES-SO Valais TechnoArk 3 CH-3960 Sierre, Switzerland 0 0 1 44 253 university of geneva 2 1 296 14.0 Normal 0 false false false DE-AT JA X-NONE henning.mü[email protected]

References

[1] Henning Müller, Nicolas Michoux, David Bandon, Antoine Geissbuhler, A review of content-based image retrieval systems in medicine – clinical benefits and future directions, International Journal of Medical Informatics volume 73, pages 1-23, 2004. [2] Daniel Rubin, Creating and curating a terminology for radiology: ontology modeling and analysis, Journal of Digital Imaging, Volume 21, Issue 4, pages 355-362, 2008 [3] Pragya Dang, Mannudeep Kalra, Thomas Schulz, Steven Graham, Keith Dreyer: Render: An Online Searchable Radiology Study Repository, RadioGraphics 2009; 29: 1233-1246. [4] Axel Gerstmair, Philipp...