Related Experiment Video
Updated: Jun 7, 2025
07:44
Evaluation of Patients' Posture and Gait Profile After Lumbar Fusion Surgery by Video Rasterstereography and Treadmill Gait Analysis
Published on: March 23, 2019
17.7K
Predicting functional outcome after open lumbar fusion surgery: A retrospective multicenter cohort study
1Department of General Surgery, Suzhou Ninth Hospital Affiliated to Soochow University, Suzhou, Jiangsu Province, China; Department of Radiology, Suzhou Ninth Hospital Affiliated to Soochow University, Suzhou, Jiangsu Province, China.
European Journal of Radiology
|November 18, 2024
Summary
A new model accurately predicts functional recovery after lumbar fusion surgery using radiomics and clinical data. This tool aids surgeons in determining surgical necessity for improved patient outcomes.
Area of Science:
- Spine Surgery
- Radiology
- Medical Informatics
Background:
- Lumbar fusion surgery aims to improve function, but outcomes vary.
- Predicting short-term functional outcomes is crucial for patient selection and management.
- Existing prediction tools may lack accuracy or generalizability.
Purpose of the Study:
- To develop and externally validate a predictive tool for short-term functional outcomes post-lumbar fusion.
- To integrate radiomics features with clinical data for enhanced predictive power.
- To assess the clinical utility and stability of the developed model.
Main Methods:
- Analysis of 1520 patients undergoing lumbar fusion across three institutions.
- Extraction of radiomics features from paraspinal muscles using CT and MRI scans.
- Development of a combined predictive model using multivariable logistic regression, integrating radiomics scores and clinical factors.
- External validation and performance evaluation using decision and calibration curves, with SHAP for interpretation.
Main Results:
- A combined model incorporating radiomics score and four clinical features demonstrated excellent predictive performance.
- The model achieved high AUC values (0.79-0.85) across derivation and independent test datasets.
- The combined model significantly outperformed clinical models and radiomics alone, showing superior calibration and utility.
Conclusions:
- The developed combined model accurately predicts functional outcomes after lumbar fusion.
- This tool can assist clinicians in decision-making regarding surgical necessity and expected functional recovery.
- Radiomics integration offers a promising approach for personalized surgical planning in spine fusion.
