Predicting Spinal Cord Injury Prognosis Using Machine Learning: A Systematic Review and Meta-Analysis
ABSTRACT
Background:
Spinal cord injury is complicated and varied conditions that receive a lot of attention. However, the prognosis of patients with spinal cord injury is increasingly being predicted using machine learning techniques.
Objective:
The purpose of this study is to evaluate the efficacy and caliber of machine learning models in forecasting the consequences of spinal cord injury.
Methods:
Literature searches were conducted in PubMed, Web of Science, Embase, PROSPERO, Scopus, Cochrane Library, CNKI, CBM, and Wanfang databases. Meta-analysis of the area under the receiver operating characteristic curve (AUC) of machine learning models was performed to comprehensively evaluate their performance.
Results:
A total of 1254 articles were retrieved, and 13 eligible studies were included. Predictive outcomes included spinal cord function prognosis, postoperative complications, independent living ability, and walking ability. For spinal cord function prognosis, the AUC of the Random Forest (RF) algorithm was 0.832, the AUC of the Logistic Regression (LR) algorithm was 0.813 (95% CI: 0.805, 0.883), the AUC of the Decision Tree (DT) algorithm was 0.747 (95% CI: 0.677, 0.802), and the AUC of the XGBoost algorithm was 0.867. For postoperative complications, the AUC of the RF algorithm was 0.627 (95% CI: 0.441, 0.812), the AUC of the LR algorithm was 0.747 (95% CI: 0.597, 0.896), and the AUC of the DT algorithm was 0.688. For independent living ability, the AUC of the CART model was 0.813. For walking ability, the model based on the VM algorithm was the most effective, with an AUC of 0.780.
Conclusions:
Machine learning models predict spinal cord injury outcomes with relative accuracy, particularly in spinal cord function prognosis. They are expected to become important tools for clinicians in assessing the prognosis of spinal cord injury patients, with the XGBoost algorithm showing the best performance. Prediction models should continue to advance as large data is used and machine learning algorithms develop.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.