A Prospective Controlled Trial of Large Language Model–based Simplification of Oncologic CT Reports for Patients with Cancer
Philipp Prucker, Keno K. Bressem, Jan C. Peeken, Mateo Jukic, Alexander W. Marka, Maximilian Strenzke, Su Hwan Kim, Christian Mertens, Dominik Weller, Tristan Lemke, Markus Graf, Sebastian Ziegelmayer, Avan Kader, Jacqueline Lammert, Mat Makowski, Felix Busch, Lisa C. Adams
Abstract
< .001). Radiologist review revealed factual errors in 6% (moderate, 2%; severe, 4%), content omissions in 7% (minor, 2%; moderate, 1%; severe, 4%), and inappropriate additions in 3% (minor, 1%; moderate, 2%) of simplified reports. Conclusion LLM simplification of oncologic CT reports improved patient comprehension and reduced reading burden. However, clinically relevant errors were identified. © RSNA, 2025
Topics & Concepts
MedicineReadabilityLikert scaleLogistic regressionComprehensionCognitionMedical physicsWorkloadRadiologyReading (process)Odds ratioProspective cohort studyRandomized controlled trialCancerClinical trialMEDLINEMedical recordOddsReading comprehensionScale (ratio)Physical therapyRadiology practices and educationArtificial Intelligence in Healthcare and EducationText Readability and Simplification