The Potential Role of Artificial Intelligence in Systematic Follow-Up Recommendation Tracking and Outcome Assessment

Area of Science:

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

Background:

• Actionable findings requiring follow-up are common in radiology reports.
• Inconsistent completion of recommended follow-up can delay diagnoses, including cancer.
• Current follow-up tracking systems require significant manual effort.

Purpose of the Study:

• To review the components of radiology follow-up recommendation management systems.
• To explore the potential of artificial intelligence (AI) in improving these systems.
• To enhance efficiency and expand capabilities in managing diagnostic follow-up.

Main Methods:

• Review of key components: identification, communication, tracking, and outcomes.
• Exploration of AI, specifically large language models, for data extraction and aggregation.
• Discussion of how AI can optimize manual processes in follow-up management.

Main Results:

• AI offers potential for automating data extraction from unstructured radiology reports.
• Large language models can improve efficiency in managing follow-up recommendations.
• AI can enhance the robustness and completeness of follow-up tracking systems.

Conclusions:

• AI integration can significantly improve the efficiency of radiology follow-up management.
• AI tools can help ensure completion of recommended diagnostic procedures.
• Implementing AI can lead to better patient outcomes by closing the care loop.