AI Matches Expert Readers for Longitudinal Scoliosis Monitoring on EOS Radiographs

An AI tool showed near-perfect agreement (ICC 0.97-0.99) with expert readers for Cobb angle measurements on 144 pediatric EOS spine radiographs, identifying ≥5° progression with 91% accuracy. Mean differences were negligible (0.2°-0.7°), supporting consistent, efficient scoliosi…

• Single-center retrospective study of 144 EOS radiographs from 29 scoliosis patients (median age 13). AI was IB Lab SQUIRREL software.
• AI measurement error showed no significant temporal drift across sequential exams (all p > 0.05), indicating reliable longitudinal performance.
• Limitation: Retrospective, single-center design; software was not validated on an external independent cohort.

RadPigeon summaries are original and for information only. They are not clinical advice.