EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

Stability‐Driven Selection of EEG Connectivity Features for Psychosis Classification: A Network‐Based Machine Learning Approach

Mahdi Naeim, Mohammad Narimani iD

CNS Neuroscience & Therapeutics · 2026-05 · 32 (5)

DOI: 10.1002/cns.70943 ↗

At a glance

Journal article in CNS Neuroscience & Therapeutics, published online on 2026-05-20. Written by Mahdi Naeim and 1 co-authors. Institution: Department of Psychology, Faculty of Educational Sciences and Psychology, University of Mohaghegh Ardabili, Ardabil, Iran. Major MEDLINE topics: Electroencephalography, Psychotic Disorders, Machine Learning, Brain. Cited 8 times so far (Crossref); 25 references. The full text is freely available (open access, CC BY 4.0).

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Objectives

To develop a network‐based machine learning framework for classifying psychosis using EEG connectivity features and to identify stable, reproducible candidate markers through a stability‐driven approach.

Methods

This study was designed as a cross‐sectional analytical secondary data analysis using a publicly available EEG dataset comprising 43 participants (19 psychosis, 24 controls). Functional connectivity measures (PLV and coherence) and graph‐theoretical network features were extracted across frequency bands. A nested cross‐validation framework with subject‐wise splitting was applied. Feature importance was evaluated using permutation importance and SHAP, and stability scores were computed across folds to identify robust features. Classification was performed using support vector machine (SVM) and random forest (RF) models.

Results

The SVM model achieved superior performance (accuracy: 91.2%, AUC: 0.967). Stability analysis identified theta‐band connectivity features, particularly fronto‐parietal PLV and global efficiency, as the most reliable and discriminative. SHAP analysis confirmed their consistent contribution across subjects. However, these findings should be interpreted as exploratory given the relatively small sample size and the use of a single dataset.

Conclusion

Stable connectivity‐derived network features provide interpretable and robust candidate EEG markers for psychosis classification. The proposed framework enhances reproducibility and supports the potential of EEG‐based tools for clinical screening, while emphasizing the need for external validation.

Subjects

MeSH terms (MEDLINE)

ElectroencephalographyPsychotic DisordersMachine LearningBrainHumansSupport Vector MachineCross-Sectional StudiesFemaleMaleRandom ForestYoung AdultAdult

Author keywords

EEG connectivitySHAP interpretabilitynetwork featurespsychosis classificationstability‐driven feature selection

Authors and institutions

  • Department of Psychology, Faculty of Educational Sciences and Psychology, University of Mohaghegh Ardabili, Ardabil, Iran.

ORCID profiles: Mohammad Narimani

Cite

APA

Naeim, M., & Narimani, M. (2026). Stability‐Driven Selection of EEG Connectivity Features for Psychosis Classification: A Network‐Based Machine Learning Approach. CNS Neuroscience & Therapeutics, 32(5). https://doi.org/10.1002/cns.70943

Vancouver

Naeim M, Narimani M. Stability‐Driven Selection of EEG Connectivity Features for Psychosis Classification: A Network‐Based Machine Learning Approach. CNS Neuroscience & Therapeutics. 2026;32(5). doi:10.1002/cns.70943

BibTeX
@article{Naeim2026,
  author = {Naeim, Mahdi and Narimani, Mohammad},
  title = {Stability‐Driven Selection of EEG Connectivity Features for Psychosis Classification: A Network‐Based Machine Learning Approach},
  journal = {CNS Neuroscience & Therapeutics},
  year = {2026},
  volume = {32},
  number = {5},
  doi = {10.1002/cns.70943}
}
Start for free →