Reducing Diagnosis Error in Radiology - Is It Possible? | Domain: Radiologist Performance Improvement (2024)-M8-RCP24-2024

Video Transcription

Video Summary

The RSNA session discussed strategies for reducing diagnostic errors in radiology, exploring tools like driver diagrams and models for improved patient safety. The driver diagram aids in identifying primary and secondary causes of diagnostic delays to propose effective interventions. Key components contributing to errors include missed findings, misinterpretation, and communication gaps, with interventions focusing on raising awareness and utilizing AI. Cognitive errors like framing, availability bias, and premature closure contribute significantly to radiology diagnostic errors. Structured reporting, standardized terminology, and improved patient history documentation are recommended to mitigate these biases.<br /><br />The session underscored the importance of diagnostic certainty in reports, avoiding vague terms like "perhaps" or "possibly." A standardized diagnostic certainty scale can enhance communication and actionability of radiology reports, fostering referrer trust. Attendees discussed AI's potential to enhance diagnostic accuracy by minimizing variability, improving image quality, and aiding in standardizing reporting. Challenges remain, including whether AI should retrospectively assess past diagnostic errors, with implications for medical legal concerns and radiologist performance assessment. Overall, the session showcased strategies and emerging technologies aimed at enhancing radiological diagnostics through consistency, accuracy, and improved communication.

Keywords

diagnostic errors

radiology

driver diagrams

patient safety

AI in radiology

cognitive errors

structured reporting

diagnostic certainty

standardized terminology