User Interface Design for AI-Based Medical Imaging: A Scoping Review

Area of Science:

• Medical Imaging and Artificial Intelligence
• Human-Computer Interaction
• Clinical Informatics

Background:

• Graphical User Interfaces (GUIs) are crucial for computer-aided diagnosis (CAD) systems in medical imaging.
• A gap exists in literature regarding GUI design and clinical workflow integration for CAD.
• Effective GUIs enhance usability, interpretability, and adoption of AI-driven medical imaging tools.

Purpose of the Study:

• To review and analyze GUI design approaches and trends in CAD systems for medical image analysis.
• To focus on classification and segmentation tasks within CAD systems.
• To examine interaction patterns, clinical applications, datasets, AI algorithms, and evaluation protocols.

Main Methods:

• A scoping review of scientific publications from 2020-2024 across five major databases.
• Used search string: "medical image" AND ("computer-aided diagnosis" OR "CADx" OR "CAD") AND "user interface".
• Included 46 studies after screening and applying inclusion/exclusion criteria.

Main Results:

• Growing trend of integrating interactive machine learning, uncertainty visualization, and explainable AI (XAI) into GUIs.
• CAD system GUIs are evolving from static displays to interactive, collaborative platforms.
• GUIs now offer tools for clinicians to understand, question, and refine AI analyses.

Conclusions:

• The design of GUIs for medical CAD systems is advancing towards more interactive and informative interfaces.
• There is a need for standardized usability and human-computer interaction metrics for evaluating these systems.
• Future research should focus on developing and validating these standardized evaluation metrics.