Validating the semantics of a medical iconic language using ontological reasoning

Area of Science:

• Medical Informatics
• Ontology Engineering
• Knowledge Representation

Background:

• Clinicians need efficient ways to interpret complex medical texts like guidelines and drug monographs.
• The Visual Clinical Medical (VCM) language uses icons for medical concepts but lacks formalized semantics, leading to inconsistent icon creation.
• Inconsistent icons, such as a 'tumor of sleeping' icon, hinder clear communication and accurate understanding.

Purpose of the Study:

• To formalize the semantics of the VCM language using Description Logics (DLs) and the Web Ontology Language (OWL).
• To develop an ontology for evaluating VCM icon consistency using automated reasoners.
• To address the challenge of inconsistent icon generation within the VCM language.

Main Methods:

• Designed a VCM ontology using the Protégé editor.
• Developed scripts to translate the VCM lexicon into OWL format.
• Evaluated the ontology's ability to determine icon consistency on a set of 100 random icons.

Main Results:

• The developed ontology demonstrated good performance in determining VCM icon consistency.
• The evaluation achieved high sensitivity in identifying inconsistent icons.
• The semantic formalization successfully addressed the ambiguity in VCM icon creation.

Conclusions:

• The formalized VCM ontology effectively determines icon consistency, enhancing the reliability of the iconic language.
• The ontology facilitates mapping VCM to other medical terminologies and generating textual labels for icons.
• This semantic framework supports the development of user-friendly interfaces for creating consistent VCM icons.