Informatics in radiology: an information model of the DICOM standard

Area of Science:

• Medical Informatics
• Ontology Engineering

Background:

• The Digital Imaging and Communications in Medicine (DICOM) standard is crucial for radiology but presents implementation challenges due to its complexity and reliance on human interpretation.
• Nonstandard implementations of the DICOM specification hinder seamless information system development.

Purpose of the Study:

• To create a formally sound and computationally accessible information model of the DICOM Standard.
• To address the challenges posed by the complexity and human-interpretability of the current DICOM specification for information system developers.

Main Methods:

• The DICOM Standard was modeled as an ontology, providing a machine-accessible and human-interpretable representation.
• The ontology was developed using information-modeling tools.
• It includes a real-world model (patients, studies, images) and a DICOM entity model (connections between real-world and DICOM entities).

Main Results:

• A DICOM Ontology was successfully created, offering a structured and computable representation of the standard.
• The ontology facilitates the description of real-world medical imaging features and their relation to DICOM information entities.
• This model supports the Cancer Biomedical Informatics Grid (caBIG) initiative.

Conclusions:

• The developed DICOM Ontology simplifies the DICOM standard for information system developers.
• It can be extended to cover the entire DICOM Standard.
• The ontology serves as a foundation for advanced medical imaging systems in research and clinical practice.