ShRed ML model predicts shoulder redislocation with moderate accuracy

A random forest model using MRI cartilage thickness and clinical variables predicted shoulder redislocation with 79% cross-validated accuracy (AUC 0.84) in a prospective 42-patient cohort. External validation needed.

• Prospective, single-center feasibility study in 42 patients (54.8% redislocation rate).
• Random forest outperformed five other algorithms; years since first dislocation was the strongest predictor, followed by age and glenoid cartilage thickness.
• Accuracy 95% CI 69%–89%; no external validation performed.

Abstract only; full methods, potential overfitting, and reproducibility not assessed.

Automated summary

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