The Imaging Problem List as an Accelerator for Radiology AI Applications

Radiology has a memory problem. Every imaging study generates a report, but these reports exist as isolated documents, disconnected from each other and from the patient’s broader clinical story. To interpret a new exam, a radiologist must manually reconstruct the patient’s imaging history from prior reports — piecing together which findings are new, which are stable, and which have changed. This cognitive burden consumes time and creates opportunities for important findings to be overlooked. The Imaging Problem List (IPL) solves this by giving radiology a structured, longitudinal memory: rather than burying findings as prose in documents, it extracts each observation as a discrete, trackable entity: anchored to standardized anatomy, tagged with its temporal status, and linked to the same finding across prior studies. An aortic aneurysm measured at 3.2, 3.5, 3.9, and 4.3 cm across successive years becomes a single entry with a complete trajectory visible at the point of care. 

This structured foundation is also what generative AI needs to become genuinely useful in radiology. Today’s AI tools can summarize a report but struggle to reason about how a patient’s imaging findings have evolved over years. By converting radiology’s narrative archive into structured, queryable knowledge, the IPL enables AI systems to understand context, track trajectories, and generate insights grounded in complete patient histories. This webinar introduces the IPL concept, its technical architecture, and early real-world implementations, and shows why it is the enabling infrastructure for AI to transform radiology into proactive, longitudinal patient care. 

Objectives    

1. Describe the “memory problem” in radiology reporting and its impact on radiologist efficiency and patient safety. 
2. Explain the Imaging Problem List architecture — findings extracted as discrete, anatomy-anchored, temporally-classified entities linked longitudinally across studies. 
3. Demonstrate how a structured longitudinal record supports point-of-care decisions such as incidental-finding surveillance and chronic-disease monitoring. 
4. Discuss how the IPL provides the structured foundation that enables generative AI to reason about patient trajectories and produce context-aware outputs. 

Resources

Imaging Problem List

Imaging Problem List Profile  

2026 Webinar: What Is Multimodal Healthcare Data Governance and Why Does It Matter Now?

2025 Webinar: Global Health AI for Radiology

CE Credit

The Society for Imaging Informatics in Medicine has approved this activity for 1.0 hours of SIIM IIP Credits towards certification and re-certification by the American Board of Imaging Informatics (ABII).