Evaluation of Visualisation Techniques for Meaningful Representation of Clinical Classification Data Sets

Area of Science:

• Health Informatics
• Data Visualization
• Clinical Terminology

Background:

• Clinical classification systems, such as the International Classification of Disease, version 10 (ICD-10), are transitioning from traditional paper-based formats to digital systems.
• Digital systems facilitate open processes and collaborative development of these extensive, hierarchically organized terminologies.
• Visualizing large, complex hierarchical data presents significant challenges in clinical informatics.

Purpose of the Study:

• To evaluate the suitability of various data visualization techniques for accommodating large clinical classification datasets.
• To identify the most effective method for representing the hierarchical and multi-parent structures inherent in systems like ICD-10 within digital environments.

Main Methods:

• Investigated a selection of visualization technologies, including tree views, tree view diagrams, and force-directed graphs.
• Assessed the capacity of each technique to handle the scale and structural complexity of clinical classification data.
• Focused on suitability for digital systems and collaborative processes.

Main Results:

• Tree view diagrams demonstrated superior suitability for visualizing large clinical classification datasets.
• This technique effectively accommodates the multi-parent structures found in some clinical classifications.
• Tree view diagrams offer a visually appealing representation for extensive data volumes.

Conclusions:

• Tree view diagrams are the recommended visualization technique for digital clinical classification systems.
• This method addresses the challenges of scale and structural complexity, enhancing usability and accessibility.
• Adoption of tree view diagrams can support the collaborative and open development of clinical terminologies.