EUVIMEDDescoperă EUVIMED
Systematische ÜbersichtOpen access

Machine Learning–Based Prognostic Models for Functional Outcomes in Spinal Cord Injury: Systematic Review

Machine-Learning-basierte Prognosemodelle für funktionelle Outcomes bei Querschnittlähmung: systematische Übersichtsarbeit

Yuan Liu iD, Xiangxia Meng iD, Yi Ding iD, Ruifa Yao iD, Shuchang Xu iD

JMIR Medical Informatics · 2026-06-23 · 14 · e84980

DOI: 10.2196/84980 ↗

Auf einen Blick

Die Übersichtsarbeit zeigt für 19 Kohortenstudien zu Machine-Learning-Prognosemodellen bei Querschnittlähmung eine hohe Verzerrungsgefahr und unvollständige Berichterstattung. Externe Validierung war selten.

Fragestellung

Es wurde beurteilt, wie gut Machine-Learning-basierte Prognosemodelle bei Querschnittlähmung berichtet sind, wie hoch das Bias-Risiko ist und wie klinisch anwendbar sie sind.

Methode

Systematische Übersichtsarbeit; Suche in CNKI, Wanfang Data, VIP Database, Sinomed, PubMed, Web of Science, Embase und Scopus bis 14. Mai 2025; Bewertung mit TRIPOD und PROBAST.

Ergebnisse

19 Kohortenstudien. TRIPOD-Übereinstimmung 54,8 % (17/31) bis 81,1 % (30/37), Median 74,2 % (IQR 64,5 %–77,4 %). Alle 19 Studien mit hohem Verzerrungsrisiko; 1 Studie mit externer Validierung.

Einschränkungen

Keine Studienschätzung der Stichprobengröße; Kalibrierung selten berichtet; große Heterogenität; externe Validierung nur in 1 Studie.

Schlussfolgerung der Autor:innen

Machine Learning zeigt Potenzial für Prognosemodelle bei Querschnittlähmung. Die Evidenz ist jedoch durch unvollständige Berichterstattung, hohes Bias-Risiko, Heterogenität und wenig externe Validierung begrenzt.

Kernaussagen

  • 19 Kohortenstudien wurden eingeschlossen
  • Alle 19 Studien hatten ein hohes Verzerrungsrisiko
  • Externe Validierung wurde nur in 1 Studie berichtet

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

Systematische Übersicht in JMIR Medical Informatics, published online on 2026-06-23. Written by Yuan Liu and 4 co-authors. Institution: College of Nursing, Shandong University of Traditional Chinese Medicine, Jinan, China. Major MEDLINE topics: Spinal Cord Injuries, Machine Learning. Cited 1 times so far (Crossref); 69 references. The full text is freely available in PubMed Central.

Auf einen BlickRezumatSubjectsDate bibliograficeCiteRelated studies

Abstract (original language)

Background

Machine learning is increasingly used to develop prognostic prediction models for spinal cord injury. Nevertheless, current studies exhibit heterogeneity in outcome measures, predictors, modeling strategies, and validation methods. Moreover, the reporting quality, risk of bias, and clinical applicability of these models have not been systematically evaluated using assessment tools specific to prediction models.

Objective

This review aimed to assess the reporting quality and risk of bias of machine learning–based prognostic models for spinal cord injury, and evaluate their clinical applicability, model features, validation, and implementation barriers.

Methods

We searched the China National Knowledge Infrastructure (CNKI), Wanfang Data, VIP Database, Sinomed, PubMed, Web of Science, Embase, and Scopus databases from their inception up to May 14, 2025. Two investigators independently screened studies, extracted data, and assessed risk of bias. Reporting quality and risk of bias were evaluated using the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) statement and the Prediction Model Risk of Bias Assessment Tool (PROBAST). Descriptive statistics and narrative synthesis were used to summarize the findings.

Results

In total, 19 cohort studies were included. TRIPOD adherence ranged from 54.8% (17/31) to 81.1% (30/37), with a median of 74.2% (IQR 64.5%-77.4%). Overall, all 19 studies were judged to have a high risk of bias, mainly because of limitations in the analysis domain. Only 1 (5.3%) study included external validation, while 16 (84.2%) studies used internal validation and 2 (10.5%) studies reported model development only. No study justified the sample size; 6 (31.6%) studies reported imputation or other methods for handling missing data, and calibration was rarely reported.

Conclusions

Machine learning shows potential for spinal cord injury prognostic modeling, especially when complex clinical or imaging data are available. However, existing evidence is limited by incomplete reporting, high risk of bias, substantial heterogeneity, and limited external validation. Larger, methodologically robust studies with standardized outcomes, external validation, and evaluation of clinical usefulness are necessary before these models can be implemented in routine clinical practice. Trial Registration PROSPERO CRD420251071502; https://www.crd.york.ac.uk/PROSPERO/view/1071502

Subjects

MeSH terms (MEDLINE)

Spinal Cord InjuriesMachine LearningHumansPrognosisPredictive Learning Models

Author keywords

machine learningprediction modelprognosisrisk of biasspinal cord injurysystematic review

Authors and institutions

  • College of Nursing, Shandong University of Traditional Chinese Medicine, Jinan, China.
  • Healthcare Office, The Second Affiliated Hospital of Shandong University of Traditional Chinese Medicine, Jinan, China.
  • Department of Rehabilitation Medicine, The Second Affiliated Hospital of Shandong University of Traditional Chinese Medicine, Jinan, China.

ORCID profiles: Yuan Liu, Xiangxia Meng, Yi Ding, Ruifa Yao, Shuchang Xu

Cite

APA

Liu, Y., Meng, X., Ding, Y., Yao, R., & Xu, S. (2026). Machine Learning–Based Prognostic Models for Functional Outcomes in Spinal Cord Injury: Systematic Review. JMIR Medical Informatics, 14, e84980. https://doi.org/10.2196/84980

Vancouver

Liu Y, Meng X, Ding Y, Yao R, Xu S. Machine Learning–Based Prognostic Models for Functional Outcomes in Spinal Cord Injury: Systematic Review. JMIR Medical Informatics. 2026;14:e84980. doi:10.2196/84980

BibTeX
@article{Liu2026,
  author = {Liu, Yuan and Meng, Xiangxia and Ding, Yi and Yao, Ruifa and Xu, Shuchang},
  title = {Machine Learning–Based Prognostic Models for Functional Outcomes in Spinal Cord Injury: Systematic Review},
  journal = {JMIR Medical Informatics},
  year = {2026},
  volume = {14},
  pages = {e84980},
  doi = {10.2196/84980}
}
Start for free →