An interpretable and trustworthy AI framework for large-scale longitudinal structure-pain association studies using data from the Osteoarthritis Initiative (OAI)

This research develops an interpretable AI framework combining deep learning-based MRI Osteoarthritis Knee Score (MOAKS) prediction with interpretable statistical modeling to study structure-pain relationships at scale using data from the Osteoarthritis Initiative (OAI). It utilizes deep learning for MOAKS feature prediction from MRIs with uncertainty quantification, followed by a longitudinal latent class mixed model to examine associations between structural abnormalities and knee pain.