Psychological and Clinical Factors for Secondary ACL Injury Risk Prediction

487 patients who underwent ACL reconstruction were assessed across five domains to predict secondary injury risk. The model, achieving an AUC of 0.739, identified psychological factors like kinesiophobia and depressive symptoms as the most influential predictors. This retrospective cohort study by Shengjie Xiong was last updated in May 2026.

Use Cases

• Training machine learning models for secondary injury risk based on multi-domain patient data.
• Investigating the predictive contribution of psychological factors like kinesiophobia (TSK score) and depression (PHQ-9) in orthopedic recovery.
• Benchmarking logistic regression and random forest algorithms on a clinical dataset with 30 predictor variables.
• Conducting SHAP analysis for model interpretability in a medical context.
• Evaluating the clinical utility of low-threshold screening models with high negative predictive value.

Strengths

• Includes data from 487 patients across five distinct clinical domains: demographic, MRI, gait analysis, isokinetic strength, and psychological.
• Model performance is reported with specific metrics, including an AUC of 0.739 and a Brier score of 0.106.
• Analysis includes interpretability techniques like SHAP and domain ablation, confirming the largest predictive contribution from the psychological domain.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small at 19.0 KB, indicating limited scope.

Provenance

Source

Shengjie Xiong via figshare.

Collection Method

Retrospective cohort study of patients who underwent primary ACL reconstruction.

Freshness

Last updated 2026-05-15 04:28:11; freshness should be verified.