Related Experiment Video
Updated: Jun 27, 2025
09:32
Cortical Source Analysis of High-Density EEG Recordings in Children
Published on: June 30, 2014
21.3K
Optimizing feature subset for schizophrenia detection using multichannel EEG signals and rough set theory
Sridevi Srinivasan1, Shiny Duela Johnson1
1Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, India.
Cognitive Neurodynamics
|May 3, 2024
Summary
This study introduces a new method for diagnosing schizophrenia (SZ) using electroencephalogram (EEG) signals. The Crossover-boosted Archimedes optimization algorithm with rough sets for Schizophrenia detection (CAORS-SD) significantly improves diagnostic accuracy and efficiency.
Area of Science:
- Neuroscience
- Computational Psychiatry
- Biomedical Engineering
Background:
- Schizophrenia (SZ) diagnosis is challenging and time-consuming via visual assessment.
- Electroencephalogram (EEG) signals offer effective insights into brain states for SZ detection.
- Existing deep learning methods for SZ detection require substantial computational resources.
Purpose of the Study:
- To develop an efficient dimensionality reduction technique for EEG signals in SZ detection.
- To propose a novel algorithm, CAORS-SD, integrating improved CAO (ICAO) and rough sets for optimal feature selection.
- To enhance the accuracy and reduce the computational cost of SZ diagnosis using multichannel EEG data.
Main Methods:
- Multichannel EEG signals from SZ patients and healthy controls were decomposed using multivariate empirical mode decomposition into multivariate intrinsic mode functions (MIMFs).
- Various entropy metrics (spectral, permutation, approximate, sample, SVD) were calculated on the MIMF domain.
- A Crossover-boosted Archimedes optimization algorithm (AOA) with rough sets (CAORS-SD) was employed for dimensionality reduction and feature selection.
Main Results:
- The CAORS-SD model achieved high diagnostic performance: accuracy (96.34%), sensitivity (98.95%), specificity (96.86%), precision (98.52%), and F1-score (96.74%).
- The proposed method significantly reduced processing time and minimized the error rate compared to existing approaches.
- Dimensionality reduction using ICAO and rough sets optimized feature selection for the kernel support vector machine classifier.
Conclusions:
- The CAORS-SD method demonstrates superior performance in accurately detecting schizophrenia from EEG signals.
- The integration of ICAO dimensionality reduction and rough set-based feature selection enhances diagnostic efficiency.
- This approach offers a computationally efficient and highly accurate tool for schizophrenia detection, reducing overfitting risk.
