Machine learning-based prediction model for paresthesia improvement after open carpal tunnel release: a preliminary study.

BACKGROUND: Carpal tunnel syndrome (CTS) is the most common entrapment neuropathy of the hand. While carpal tunnel release surgery generally provides good outcomes, some patients continue to experience persistent paresthesia postoperatively. This study aimed to develop a machine learning-based prediction model for paresthesia improvement following carpal tunnel surgery. METHODS: We retrospectively analyzed 94 hands from 83 patients who underwent carpal tunnel release surgery between April 2021 and March 2024. Paresthesia improvement was evaluated at 1-year follow-up using a 10-point scale, with excellent improvement defined as scores 0-1. Three machine learning algorithms (Support Vector Machine, Random Forest, and Logistic Regression) were compared. Model performance was evaluated using accuracy, precision, recall, F1-score, and area under the ROC curve (AUC). RESULTS: Of 94 hands, 53 (56.4%) achieved excellent paresthesia improvement. The optimized SVM model demonstrated the best performance with an AUC of 0.852, accuracy of 78.9%, precision of 88.9%, and recall of 72.7%. The five most important predictive factors were: distal motor latency (DML), age, disease duration, thenar muscle atrophy, and thumb opposition deficit. CONCLUSIONS: Our preliminary machine learning model successfully predicted excellent paresthesia improvement after carpal tunnel surgery with good accuracy using five easily obtainable clinical parameters. This pilot study provides a foundation for developing more robust prediction tools that could assist in preoperative counseling and treatment planning. CLINICAL TRIAL REGISTRATION: Clinical trial number: not applicable. This study was a retrospective analysis and not a clinical trial.