Skip to main content
Search

# Knee injuries and reconstruction techniques

1 related articles

Subscribe to updates
  1. Suzhou University team uses tibial plateau X-ray parameters plus machine learning to identify complete and incomplete discoid lateral meniscus, reaching validation AUCs of 0.885 and 0.861

    This retrospective study of 494 patients (145 with complete discoid lateral meniscus, CDLM; 64 with incomplete discoid lateral meniscus, ICDLM; 285 normal lateral meniscus controls) measured multiple tibial plateau radiographic anatomical parameters, used LASSO to select six features (gender, lateral joint space, height of lateral tibial spine, lateral slope of the lateral tibial spine, lateral slope of the medial tibial spine, and tibial eminence width/tibial plateau width ratio), built seven machine-learning models, and evaluated them in a held-out validation set, where the best CDLM model (support vector machine) reached an AUC of 0.885 and the best ICDLM model (gradient boosting) reached an AUC of 0.861.

You’re all caught up