PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
March 8, 2026Frontiers in Neuroscience2 citationsOpen Access

Schizophrenia detection via lobe-wise and overall EEG features using VMD and bayesian-optimized machine learning models

GSGandham Sai SravanthiLSLakhan Dev Sharma

Key Result

The proposed VMD and optimized machine learning framework achieved a classification accuracy of 96.7% on the MHRC dataset and 99.0% on the RepOD dataset for detecting schizophrenia using EEG signals.

Key Points

  • The aim is to develop an efficient method for early schizophrenia detection using EEG signals and machine learning techniques.
  • Utilized two EEG datasets: MHRC and RepOD for analysis.
  • Segmented EEG signals into 8-second intervals and decomposed using Variational Mode Decomposition (VMD).
  • Extracted multi-domain features from the resulting Intrinsic Mode Functions (IMFs).
  • Classified features using nine machine learning and seven optimized machine learning classifiers.
  • Applied subject-wise Leave-One-Out Cross-Validation (LOOCV) to prevent data leakage.
  • Achieved 96.7% accuracy for MHRC using the Optimizable KNN classifier.
  • Attained 99.0% accuracy for RepOD using the Optimizable Ensemble classifier.
  • Frontal lobe analysis showed 91.2% accuracy for MHRC and 99.4% for RepOD using the Optimizable Neural Network.
  • Found strong discriminative power in temporal lobe features.

Study Design

Type

Cross-Sectional (n=112)

Multicenter

Yes

Structured PICO

Does a VMD and optimized machine learning framework improve the accuracy of schizophrenia detection using EEG signals compared to healthy controls?

P
Population
112 participants across two independent datasets (MHRC and RepOD) comprising individuals diagnosed with schizophrenia and healthy controls, evaluated for EEG-based diagnostic classification.
E
Exposure
Variational Mode Decomposition (VMD) of EEG signals into 10 Intrinsic Mode Functions (IMFs) combined with Optimized Machine Learning (OML) classifiers.
C
Comparator
Healthy controls (and baseline non-optimized machine learning models).
O
Outcome
Accuracy of schizophrenia detection.surrogate

The VMD and optimized machine learning framework provides a highly accurate and computationally efficient method for early schizophrenia detection using EEG signals.

Limitations

  • Relatively small sample sizes in the evaluated datasets (e.g., 28 participants in the RepOD dataset)
  • Cross-sectional nature of the data limits longitudinal evaluation of EEG changes over the course of the disease

Abstract

Schizophrenia (SCH) is a chronic and severe mental disorder that leads to significant cognitive and neurophysiological impairments, affecting daily life. Early diagnosis remains challenging as it relies on the manifestation of symptoms that develop over time. Electroencephalography (EEG), which measures brain activity, provides a promising avenue for early detection. In this study, two EEG datasets—the Mental Health Research Center (MHRC) and the Repository for Open Data (RepOD)—were employed to detect SCH. EEG signals were segmented into 8-second durations and decomposed using Variational Mode Decomposition (VMD) into 10 Intrinsic Mode Functions (IMFs). Multi-domain features extracted from IMFs were classified using nine machine learning (ML) and seven optimized ML (OML) classifiers. The proposed method achieved an accuracy (Ac) of 96.7% for the MHRC dataset using the Optimizable KNN classifier and 99.0% for the RepOD dataset using the Optimizable Ensemble classifier. To prevent data leakage, a strict subject-wise Leave-One-Out Cross-Validation (LOOCV) strategy was employed. Lobe-wise analysis showed that the frontal lobe achieved accuracies of 91.2% for MHRC using the Optimizable Ensemble and 99.4% for RepOD using the Optimizable Neural Network, with the temporal lobe also showing strong discriminative power. These findings align with established evidence of frontal–temporal dysconnectivity in SCH. Overall, the proposed VMD + OML framework offers a computationally efficient and clinically interpretable solution for early SCH detection using EEG signals.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Sravanthi et al. (2026) conducted a cross-sectional in Schizophrenia (n=112). Variational Mode Decomposition (VMD) and optimized machine learning (OML) on EEG signals vs. Healthy controls was evaluated on Classification accuracy for schizophrenia detection. The proposed VMD and optimized machine learning framework achieved a classification accuracy of 96.7% on the MHRC dataset and 99.0% on the RepOD dataset for detecting schizophrenia using EEG signals.

synapsesocial.com/papers/69ad122be7e9681137aa87f7https://doi.org/10.3389/fnins.2026.1753779
Ask AI
Helpful
Bookmark
Share
View Full Paper