Levels of Autonomous Radiology

Area of Science:

• Medical Imaging and Radiology
• Artificial Intelligence in Healthcare
• Health Informatics

Background:

• Radiology, a relatively young medical discipline, has undergone significant technological evolution.
• Recent decades have seen an exponential increase in medical imaging data generation.
• Artificial intelligence (AI) applications are poised to drive the next major advancement in radiology.

Purpose of the Study:

• To outline a structured, level-wise classification for the progression of automation in radiology.
• To detail AI assistance at each automation level within radiology.
• To identify and propose solutions for challenges associated with AI adoption in radiology.

Main Methods:

• Development of a hierarchical framework categorizing AI integration in radiology.
• Analysis of AI's role in automating tasks like annotation and report generation.
• Examination of AI's contribution to initial patient assessment and imaging feature analysis.

Main Results:

• A proposed level-wise classification system for AI automation in radiology.
• Identification of specific AI applications and their impact at each progression level.
• Discussion of challenges and potential solutions for implementing AI in radiological workflows.

Conclusions:

• AI integration represents the next evolutionary phase for radiology, enhancing diagnostic and treatment planning workflows.
• A structured approach to AI adoption, addressing challenges proactively, is crucial for successful implementation.
• This framework aims to guide the seamless integration of new AI technologies into modern radiology practices.