Semantic DICOM


The existing massive data repositories in worldwide healthcare are expected to contain unprecedented opportunities and gains for quality, effectiveness and cost reduction of healthcare: a treasure chest of medical data which is by now nearly untapped compared to its potential.

With SeDI we would like to overcome the difficulties tapping that treasure caused by the diversity of data silos in healthcare that have been piling up over the years.
↓ Read more …

In current clinical practice it is common that each department (e.g. radiology, histology, nuclear medicine, radiotherapy, etc) uses different silos from different vendors with different data formats. The ability to integrate and simultaneously query these heterogenous types of data will yield even more opportunities to leverage the value of the data.

The technology enabling this is called semantic web technology and is on the rise in a multitude of sectors handling ‚big data‘. In healthcare however, accurate solutions are not yet commercially available.

The opportunities in radiotherapeutic practice are huge: semantic web technologies are very promising for their use in cancer phenotyping (the characterisation of a tumour). For phenotyping, a diverse range of data needs to be integrated to optimally assess the tumour type and its extent in order to predict treatment response and enable optimal treatment planning [Lambin, Dekker, et al., Nature Reviews in Clinical Oncology. 2012; in press]. Relevant data types for tumour phenotyping include multiple functional and morphological imaging modalities as well as genotype, patient history and environmental data. Integrating all these is currently virtually impossible due to the great variety of systems and vendors as well as the high rate of improvements of the systems and is even worsened if data from multiple institutes needs to be integrated in e.g. multicentre clinical trials. Full integration would however becomes feasible with the introduction of semantic web technology.

↑ Hide.

top of page


SeDI will be the first semantic web solution for radiology with a unique approach as it does not foresee in the translation of existing databases into a semantic format, but the real-time translation of semantic queries into DICOM queries. This creates light-weight and highly flexible software, applicable to virtually any existing medical repository and thus any medical setting.

SeDI will deliver.

  • The first implemented semantic web system in radiotherapeutic practice
  • The validation of the value of semantic web technology in radiotherapeutic practice
  • The first open-source ontology for radiotherapy as well as a system in place to manage it
  • The first stand-alone semantic PACS system
  • The first semantic adapter to expose conventional PACS installations as semantic PACS system

↓ Read more …

The current standard in radiological practice is DICOM. This standard is purely syntactic and does not allow for the integration of semantics. It is however highly integrated in current practice and new standards will take ages to become widely accepted. It is therefore our strategy to not change this standard or propose a new one, but build upon this standard.

↑ Hide.

top of page


Title Semantic DICOM – Development of a semantic web module and
PACS system for radiotherapeutic practice
Acronym SEDI
Area Healthcare / Data processing
Participants SOHARD Software GmbH (industrial participant)
MAASTRO clinic (scientific participant)
Start 01-May-2013
End 30-Apr-2015
ID E! 7831 SEDI

top of page

Open Source

The source code of SeDI is licensed under the AGPL version 3. Please contact us if you are interested in the source code.

top of page


Logo MAASTRO clinic MAASTRO clinic (Maastricht Radiation Oncology clinic) provides patient care with radiotherapy and performs clinical radiotherapy research.
Logo der SOHARD Software GmbH SOHARD Software contributes its expertise in all database aspects related to DICOM and similar formats as well as semantic web technology.

top of page