- Reported high diagnostic accuracy (mean 93.38%) likely inflated by methodological heterogeneity and artefacts, undermining inferential validity.
- Deep learning outperformed classical machine learning (96.83% vs 87.68%), but external validation scarce (4.7%) and data leakage evident from segment-wise versus subject-wise results.
- PROBAST+AI assessment found 76.7% studies at high risk of bias; current evidence insufficient for clinical implementation, recommending standardised validation and optimised methodology.
Eur Psychiatry. 2026 Oct 15;69(1):e101. doi: 10.1192/j.eurpsy.2026.12262.
ABSTRACT
BACKGROUND: The application of artificial intelligence (AI) to resting-state electroencephalography (EEG) data represents a burgeoning frontier in schizophrenia diagnostics. However, the inferential validity of findings is frequently compromised by methodological heterogeneity. This systematic review rigorously evaluates the diagnostic efficacy and methodological robustness of AI-generated EEG biomarkers by implementing the Prediction model Risk Of Bias Assessment Tool for Artificial Intelligence (PROBAST+AI).
METHODS: Following PRISMA guidelines, a systematic search of PubMed, Scopus, and Web of Science identified 43 primary studies encompassing 3,420 patients with schizophrenia or first-episode psychosis and 3,164 healthy controls (total N = 6,584). Methodological quality and validation strategies were assessed using the 34 signaling questions of the PROBAST+AI framework.
RESULTS: The aggregate mean diagnostic accuracy across the literature was 93.38%, with deep learning architectures demonstrating superior metrics relative to classical machine learning (96.83% vs. 87.68%). Critically, external validation was restricted to 4.7% of the included studies. A pronounced performance discrepancy was observed between subject-wise (88.22%) and segment-wise (98.14%) validation protocols, indicating pervasive data leakage. Overall, 76.7% of the analyzed studies were characterized by a high risk of bias, predominantly stemming from suboptimal data partitioning and inadequate independent verification.
CONCLUSIONS: Contemporary diagnostic accuracies appear to be artifactually elevated due to structural methodological deficiencies and data leakage. Consequently, current evidence does not support the immediate clinical implementation of AI-EEG biomarkers in schizophrenia. To facilitate translational progress, we propose a standardized methodological framework optimizing validation strategies.
PMID:42878218 | DOI:10.1192/j.eurpsy.2026.12262
Share Evidence Blueprint
Save to Google Notes
Search Google Scholar
Save as PDF
⭐ My Revision List
