Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 11, 2026Annals of General PsychiatryOpen Access

Diagnostic accuracy of machine learning approaches for suicide‑related outcomes: a meta‑analysis

View Full Paper
Ask AI
Bookmark
Share

Authors

PKParisa KohnepoushiMAMaryam AfraieHRHamza Rahmani

Discussion

Loading...

Member takes

Overview

Meta-analysis demonstrates machine learning's effectiveness in predicting suicide-related outcomes, suggesting clinical viability.

Key Points

  • This analysis aims to evaluate the diagnostic accuracy of machine learning models for predicting suicide-related outcomes.
  • Conducted a systematic search in multiple databases for studies published from 2010 to 2024.
  • Included studies with at least 100 participants reporting diagnostic performance metrics for machine learning models.
  • Calculated pooled estimates using a bivariate random-effects model.
  • Ensemble models showed the highest pooled AUC at 0.95 (95% CI, 0.92–0.96) along with a specificity of 0.97 (95% CI, 0.95–0.98).
  • Sensitivity for ensemble models was 0.50 (95% CI, 0.29–0.71).
  • Post-test probabilities for positive results ranged from 64% (Logistic Regression) to 88% (Ensemble Models) at a pre-test probability of 25%.

Cite This Study

Kohnepoushi et al. (2026) studied this question.

synapsesocial.com/papers/6a0172813a9f334c28272be7https://doi.org/10.1186/s12991-026-00671-4
View Full Paper
Ask AI
Bookmark
Share