EUVIMEDDescubrir EUVIMED
Systematische ÜbersichtMetaanalyseOpen access

Predictive Performance of Machine Learning for Suicide in Adolescents: Systematic Review and Meta-Analysis

Vorhersageleistung von Machine Learning für Suizid bei Jugendlichen: Systematische Übersicht und Metaanalyse

Lingjiang Liu iD, Zhiyuan Li iD, Yaxin Hu iD, Chunyou Li iD, Shuhan He iD, Shibei Zhang iD, Jie Gao iD, Huaiyi Zhu iD, Guoping Huang iD

Journal of Medical Internet Research · 2025-06-16 · 27 · e73052

DOI: 10.2196/73052 ↗

Auf einen Blick

42 Studien zeigen insgesamt eine vielversprechende Vorhersageleistung von ML-Modellen für suizidbezogene Verhaltensweisen bei Jugendlichen, besonders für Suizidversuche.

Fragestellung

Wie gut sagen ML-basierte Modelle suizidbezogene Verhaltensweisen bei Jugendlichen vorher?

Methode

Systematische Übersicht und Metaanalyse; PubMed, Embase, Cochrane und Web of Science bis 20. April 2024; Biasbewertung mit einem multivariaten Vorhersagemodell.

Ergebnisse

42 Studien mit 104 ML-Modellen und 1.408.375 Jugendlichen im Alter von 11 bis 20 Jahren. AUC: 0,79 für NSSI, 0,77 für Suizidgedanken, 0,84 für Suizidversuche, 0,82 für Suizidversuche plus Suizidgedanken, 0,75 für Suizidversuche plus NSSI.

Einschränkungen

Überwiegend interne Validierung; nur wenige Studien mit externer Validierung, was die Übertragbarkeit einschränken kann.

Schlussfolgerung der Autor:innen

ML-Techniken zeigten eine vielversprechende Vorhersageleistung für Suizidrisiko bei Jugendlichen, besonders für Suizidversuche. Ensemble-Methoden wie Random Forest und XGBoost schnitten über mehrere Endpunkte besser ab.

Kernaussagen

  • AUC für Suizidversuche: 0,84 (95 %-KI 0,83–0,86)
  • Höchste Sensitivität bei Suizidversuchen: 0,80 (95 %-KI 0,75–0,84)
  • Höchste Spezifität bei NSSI: 0,96 (95 %-KI 0,94–0,99)

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 Journal of Medical Internet Research, published online on 2025-06-16. Written by Lingjiang Liu and 8 co-authors. Institution: Department of Psychiatry, North Sichuan Medical College, Nanchong, China. Major MEDLINE topics: Machine Learning, Suicide. Cited 26 times so far (Crossref); 89 references. The full text is freely available in PubMed Central.

Auf einen BlickResumenSubjectsDatos bibliográficosCiteRelated studies

Abstract (original language)

Background

In the context of escalating global mental health challenges, adolescent suicide has become a critical public health concern. In current clinical practices, considerable challenges are encountered in the early identification of suicide risk, as traditional assessment tools demonstrate limited predictive accuracy. Recent advancements in machine learning (ML) present promising solutions for risk prediction. However, comprehensive evaluations of their efficacy in adolescent populations remain insufficient.

Objective

This study systematically assessed the performance of ML-based prediction models across various suicide-related behaviors in adolescents, aiming to establish an evidence-based foundation for the development of clinically applicable risk assessment tools.

Methods

This review assessed ML for predicting adolescent suicide–related behaviors. PubMed, Embase, Cochrane, and Web of Science databases were rigorously searched until April 20, 2024, and a multivariate prediction model was employed to assess the risk of bias. The c-index was used as the primary outcome measure to conduct a meta-analysis on nonsuicidal self-injury (NSSI), suicidal ideation, suicide attempts, suicide attempts combined with suicidal ideation, and suicide attempts combined with NSSI, evaluating their accuracy in the validation set.

Results

A total of 42 studies published from 2018 to 2024 were included, encompassing 104 distinct ML models and 1,408,375 adolescents aged 11 to 20 years. The combined area under the receiver operating characteristic curve values for ML models in predicting NSSI, suicidal ideation, suicide attempts, suicide attempts combined with suicidal ideation, and suicide attempts combined with NSSI were 0.79 (95% CI 0.72-0.86), 0.77 (95% CI 0.71-0.83), 0.84 (95% CI 0.83-0.86), 0.82 (95% CI 0.79-0.84), and 0.75 (95% CI 0.73-0.76), respectively. The ML models demonstrated the highest combined sensitivity for suicide attempt prediction, with a value of 0.80 (95% CI 0.75-0.84), and the highest combined specificity for NSSI prediction, with a value of 0.96 (95% CI 0.94-0.99).

Conclusions

Our findings suggest that ML techniques exhibit promising predictive performance for forecasting suicide risk in adolescents, particularly in predicting suicide attempts. Notably, ensemble methods, such as random forest and extreme gradient boosting, showed superior performance across multiple outcome types. However, this study has several limitations, including the predominance of internal validation methods employed in the included literature, with few studies employing external validation, which may limit the generalizability of the results. Future research should incorporate larger and more diverse datasets and conduct external validation to improve the prediction capability of these models, ultimately contributing to the development of ML-based adolescent suicide risk prediction tools.

Subjects

MeSH terms (MEDLINE)

Machine LearningSuicideHumansAdolescentSuicidal IdeationRisk AssessmentSuicide, Attempted

Author keywords

adolescent mental healthmachine learningmeta-analysispredictive modelsuicide predictionsuicide prevention

Authors and institutions

  • Department of Psychiatry, North Sichuan Medical College, Nanchong, China.
  • Sichuan Mental Health Center, Department of Psychiatry, The Third Hospital of Mianyang, Mianyang, China.
  • Department of Clinical Medicine, Southwest Medical University, Luzhou, China.

ORCID profiles: Lingjiang Liu, Zhiyuan Li, Yaxin Hu, Chunyou Li, Shuhan He, Shibei Zhang, Jie Gao, Huaiyi Zhu, Guoping Huang

Cite

APA

Liu, L., Li, Z., Hu, Y., Li, C., He, S., Zhang, S., Gao, J., Zhu, H., & Huang, G. (2025). Predictive Performance of Machine Learning for Suicide in Adolescents: Systematic Review and Meta-Analysis. Journal of Medical Internet Research, 27, e73052. https://doi.org/10.2196/73052

Vancouver

Liu L, Li Z, Hu Y, Li C, He S, Zhang S, et al. Predictive Performance of Machine Learning for Suicide in Adolescents: Systematic Review and Meta-Analysis. Journal of Medical Internet Research. 2025;27:e73052. doi:10.2196/73052

BibTeX
@article{Liu2025,
  author = {Liu, Lingjiang and Li, Zhiyuan and Hu, Yaxin and Li, Chunyou and He, Shuhan and Zhang, Shibei and Gao, Jie and Zhu, Huaiyi and Huang, Guoping},
  title = {Predictive Performance of Machine Learning for Suicide in Adolescents: Systematic Review and Meta-Analysis},
  journal = {Journal of Medical Internet Research},
  year = {2025},
  volume = {27},
  pages = {e73052},
  doi = {10.2196/73052}
}
Start for free →