Evaluation of Clinical AI-Based Diagnostic Solutions - A Multiperspective, Interdisciplinary Approach

Area of Science:

• Healthcare innovation
• Medical diagnostics
• Artificial intelligence in medicine

Background:

• The increasing volume of data in healthcare necessitates efficient diagnostic solutions.
• Artificial intelligence (AI) is rapidly advancing clinical diagnostics, but integration challenges persist.
• Assessing the true value of new AI diagnostic tools within clinical workflows is complex.

Purpose of the Study:

• To develop a structured framework for integrating AI-based diagnostic tools into clinical practice.
• To guide the process from identifying clinical needs to implementing certified AI diagnostic products.
• To ensure AI solutions are systematically evaluated throughout their lifecycle.

Main Methods:

• The INTERREG Baltic Sea Region project "Clinical Artificial Intelligence-Based Diagnostics (CAIDX)" developed a comprehensive blueprint.
• A systematic evaluation approach was employed at each development stage and throughout the AI solution's lifecycle.
• Diverse stakeholder perspectives and various evaluation methodologies were incorporated.

Main Results:

• The CAIDX project successfully produced the "Clinical AI-Pathway," an end-to-end framework.
• This framework offers methodologies and tools for systematic evaluation of AI diagnostic tools.
• The framework ensures alignment with clinical needs and rigorous value assessment.

Conclusions:

• Systematic, multi-perspective evaluation is essential for successful AI diagnostics integration.
• The "Clinical AI-Pathway" provides a structured method for assessing and implementing AI solutions.
• The framework facilitates value-driven adoption and effective use of AI in clinical diagnostics.