EUVIMEDScopri EUVIMED
ÜbersichtsarbeitSystematische ÜbersichtMetaanalyseOpen access

Predicting Spinal Cord Injury Prognosis Using Machine Learning: Systematic Review and Meta-Analysis

Prognose von Rückenmarksverletzungen mit Machine-Learning: systematische Übersicht und Metaanalyse

Linxing Zhong, Qiying Huang, Hao Zhang, Liang Xue, Yehuang Chen, Jianwu Wu, Liangfeng Wei

JMIR AI · 2025-12-05 · 4 · e66233-e66233

DOI: 10.2196/66233 ↗

Auf einen Blick

13 Studien zeigen, dass Machine-Learning-Modelle die Prognose nach Rückenmarksverletzung mit relativer Genauigkeit vorhersagen, besonders die Rückenmarksfunktion. XGBoost schnitt dabei am besten ab.

Fragestellung

Wie gut sagen Machine-Learning-Modelle die Folgen einer Rückenmarksverletzung vorher?

Methode

Systematische Übersicht und Metaanalyse; Literaturrecherche in PubMed, Web of Science, Embase, PROSPERO, Scopus, Cochrane Library, China National Knowledge Infrastructure, China Biomedical Literature Service System und Wanfang.

Ergebnisse

13 eingeschlossene Studien. Für die Prognose der Rückenmarksfunktion lagen die AUC bei 0,832 (random forest), 0,813 (logistic regression; 95 %-KI 0,805-0,883), 0,747 (decision tree; 95 %-KI 0,677-0,802) und 0,867 (XGBoost).

Schlussfolgerung der Autor:innen

Die ML-Modelle sagen SCI-Ergebnisse mit relativer Genauigkeit voraus, besonders die Prognose der Rückenmarksfunktion. XGBoost zeigte die beste Leistung.

Kernaussagen

  • 13 Studien wurden eingeschlossen
  • Beste Leistung zeigte XGBoost
  • Besonders die Rückenmarksfunktion wurde gut vorhergesagt

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

At a glance

Übersichtsarbeit in JMIR AI, published online on 2025-12-05. Written by Linxing Zhong and 6 co-authors. Institution: Department of Neurosurgery, Fuzong Clinical Medical College of Fujian Medical University, 156 West Second Ring North Road, Fuzhou, 350025, China, 86 …. Cited 9 times so far (Crossref); 40 references. The full text is freely available in PubMed Central.

Auf einen BlickAbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

Background

Spinal cord injury (SCI) is complicated and varied conditions that receive a lot of attention. However, the prognosis of patients with SCI is increasingly being predicted using machine learning (ML) techniques.

Objective

This study aims to evaluate the efficacy and caliber of ML models in forecasting the consequences of SCI.

Methods

Literature searches were conducted in PubMed, Web of Science, Embase, PROSPERO, Scopus, Cochrane Library, China National Knowledge Infrastructure, China Biomedical Literature Service System, and Wanfang databases. Meta-analysis of the area under the receiver operating characteristic curve of ML 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 area under the curve (AUC) of the random forest algorithm was 0.832, the AUC of the logistic regression algorithm was 0.813 (95% CI 0.805-0.883), the AUC of the decision tree algorithm was 0.747 (95% CI 0.677-0.802), and the AUC of the XGBoost (extreme gradient boosting) algorithm was 0.867. For postoperative complications, the AUC of the random forest algorithm was 0.627 (95% CI 0.441-0.812), the AUC of the logistic regression algorithm was 0.747 (95% CI 0.597-0.896), and the AUC of the decision tree algorithm was 0.688. For independent living ability, the AUC of the classification and regression tree model was 0.813. For walking ability, the model based on the vector machine algorithm was the most effective, with an AUC of 0.780.

Conclusions

The ML models predict SCI outcomes with relative accuracy, particularly in spinal cord function prognosis. They are expected to become important tools for clinicians in assessing the prognosis of patients with SCI, with the XGBoost algorithm showing the best performance. Prediction models should continue to advance as large data are used and ML algorithms develop.

Subjects

Author keywords

PRISMAPreferred Reporting Items for Systematic Reviews and Meta-Analysesartificial intelligence in health careclinical outcomes researchpredictive analyticsspinal cord injury

Authors and institutions

  • Department of Neurosurgery, Fuzong Clinical Medical College of Fujian Medical University, 156 West Second Ring North Road, Fuzhou, 350025, China, 86 13960760177, 86 591-87640785.

Cite

APA

Zhong, L., Huang, Q., Zhang, H., Xue, L., Chen, Y., Wu, J., & Wei, L. (2025). Predicting Spinal Cord Injury Prognosis Using Machine Learning: Systematic Review and Meta-Analysis. JMIR AI, 4, e66233-e66233. https://doi.org/10.2196/66233

Vancouver

Zhong L, Huang Q, Zhang H, Xue L, Chen Y, Wu J, et al. Predicting Spinal Cord Injury Prognosis Using Machine Learning: Systematic Review and Meta-Analysis. JMIR AI. 2025;4:e66233-e66233. doi:10.2196/66233

BibTeX
@article{Zhong2025,
  author = {Zhong, Linxing and Huang, Qiying and Zhang, Hao and Xue, Liang and Chen, Yehuang and Wu, Jianwu and Wei, Liangfeng},
  title = {Predicting Spinal Cord Injury Prognosis Using Machine Learning: Systematic Review and Meta-Analysis},
  journal = {JMIR AI},
  year = {2025},
  volume = {4},
  pages = {e66233--e66233},
  doi = {10.2196/66233}
}
Start for free →