Critical Appraisal of Artificial Intelligence-Enabled Imaging Tools Using the Levels of Evidence System

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Health Informatics

Background:

• Clinical adoption of AI imaging tools necessitates rigorous evaluation of their entire lifecycle.
• Current assessment methods lack standardization, hindering objective appraisal of technical and clinical efficacy.

Purpose of the Study:

• To propose a hierarchical evidence-based evaluation system for AI-enabled imaging tools.
• To standardize the assessment of AI tools, promoting transparency and understanding of their development and decision-making processes.

Main Methods:

• Developing a hierarchical system modeled after medical levels of evidence.
• Categorizing evidence based on quality, type, and quantity for each AI tool lifecycle component.

Main Results:

• The proposed system provides a structured approach to evaluating AI imaging tools.
• It aims to identify tools with the strongest evidence for positive patient care and health outcomes.

Conclusions:

• The proposed hierarchical system is crucial for the formal validation and regulation of AI imaging applications.
• It facilitates the safe and effective deployment of AI in clinical practice by ensuring robust evidence of efficacy.