Prediction of future knee replacement in osteoarthritis may support tailored decision making, and provide a surrogate outcome marker for clinical trials of diseases modifying osteoarthritis drugs. Fully automated cartilage segmentation is generated from knee MRI, and feature are extraction to create a radiomic signature for the prediction of knee replacement within 5 -years. The radiomic signature using univariate cox regression predicted surgery with a HR of 7.5 (p = 2e-28, 95% CI 7.1 – 7.9) which was higher than established clinico-demographic multivariate prediction model with HR 5.9 (p = 2e-28, 95%CI 5.6 – 6.2).
This abstract and the presentation materials are available to members only; a login is required.