Related Experiment Video
Updated: Jan 7, 2026
14:56
An Experimental Paradigm for the Prediction of Post-Operative Pain PPOP
Published on: January 27, 2010
21.9K
Machine Learning-Based Prediction for Axial Pain Following Expansive Unilateral Open-Door Laminoplasty: A
Kelun Huang1, Sheng Li2, Yile Dai3
1Department of Orthopedics (Spine Surgery), The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, 325035, Zhejiang, China.
Pain and Therapy
|December 10, 2025
Summary
Machine learning models can predict axial pain after expansive unilateral open-door laminoplasty (ELAP). Key predictors like C7 laminoplasty and cervical kyphosis were identified to help avoid this common complication.
Area of Science:
- Neurosurgery
- Orthopedics
- Medical Artificial Intelligence
Background:
- Axial pain is a frequent complication after expansive unilateral open-door laminoplasty (ELAP).
- Traditional statistical methods struggle to predict post-ELAP axial pain effectively.
- This study aimed to develop machine learning (ML) models for predicting axial pain and identifying its key predictors.
Purpose of the Study:
- To develop and validate machine learning models for predicting axial pain following ELAP.
- To identify significant predictors contributing to post-ELAP axial pain using SHapley Additive exPlanations (SHAP).
Main Methods:
- Retrospective analysis of 851 patients with cervical spondylotic myelopathy (CSM) undergoing ELAP.
- Feature selection using Lasso regression, followed by ML model development (XGBoost) with hyperparameter optimization.
- Temporal validation and SHAP analysis for predictor importance assessment.
Main Results:
- The extreme gradient boosting (XGBoost) model demonstrated high performance in both internal (AUC=0.948) and temporal (AUC=0.906) validation.
- Key predictors identified include C7 laminoplasty, surgical segment classification, cervical kyphosis, open-door angle, cervical lordosis, and spinal canal occupancy rate.
- SHAP analysis ranked C7 laminoplasty as the most significant predictor.
Conclusions:
- Machine learning models, particularly XGBoost, effectively predict post-ELAP axial pain.
- Identifying and addressing key predictors such as C7 laminoplasty, surgical extent, and cervical alignment can help mitigate axial pain.
- Recommendations include segment-selective ELAP, avoiding unnecessary C7 laminoplasty, and maintaining optimal open-door angles.
