A framework for data-driven adaptive GUI generation based on DICOM.

Area of Science:

• Medical Imaging
• Software Engineering
• Human-Computer Interaction

Background:

• Current diagnostic imaging software GUIs often present irrelevant tools, leading to physician cognitive overload and reduced performance.
• Existing systems lack adaptability, failing to tailor interfaces to specific clinical needs or imaging modalities.

Purpose of the Study:

• To propose a software framework for data-driven generation of adaptive GUIs in diagnostic medical imaging.
• To enhance physician efficiency and reduce visual stress by presenting only relevant functionalities.

Main Methods:

• Developed a DICOM-based mechanism for on-the-fly GUI generation, driven by image data, body part, and analysis task.
• Extended the DICOMDIR data model with a new Information Object Module (IOM) to incorporate GUI specifications.
• Integrated the framework with OsiriX imaging software as a proof-of-concept plug-in.

Main Results:

• The framework dynamically generates GUIs tailored to the specific clinical context, activating only relevant functionalities.
• The data-driven approach ensures efficient use of screen space and reduces unnecessary information for physicians.
• A proof-of-concept implementation demonstrated successful integration and adaptive GUI generation.

Conclusions:

• The proposed framework effectively addresses the limitations of static GUIs in diagnostic imaging.
• This DICOM-integrated solution enhances usability and performance in clinical workflows.
• The adaptive GUI generation holds significant potential for improving diagnostic workstations.