Related Experiment Video
Updated: Oct 25, 2025
09:35
Automatic Detection of Highly Organized Theta Oscillations in the Murine EEG
Published on: March 10, 2017
9.4K
Automated detection of schizophrenia using optimal wavelet-based norm features extracted from single-channel EEG
Manish Sharma1, U Rajendra Acharya2,3,4
1Department of Electrical and Computer Science Engineering, Institute of Infrastructure Technology Research and Management, Ahmedabad, India.
Cognitive Neurodynamics
|August 9, 2021
Summary
This study presents a computer-aided diagnosis (CAD) system using electroencephalogram (EEG) signals to accurately detect schizophrenia (SZ). The wavelet-based model achieved high accuracy, aiding clinicians in diagnosis.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Computational Psychiatry
Background:
- Schizophrenia (SZ) diagnosis relies on manual screening, which is time-consuming and error-prone.
- Electroencephalogram (EEG) signals offer a potential biomarker for neurological and psychiatric disorders.
- The nonlinear and non-stationary characteristics of EEG signals necessitate advanced signal processing techniques.
Purpose of the Study:
- To develop and validate a computer-aided diagnosis (CAD) system for schizophrenia detection.
- To leverage wavelet-based features from single-channel EEG signals for improved diagnostic accuracy.
- To provide clinicians with a reliable tool to support schizophrenia diagnosis.
Main Methods:
- Single-channel electroencephalogram (EEG) signals were collected from patients.
- Wavelet decomposition was applied to EEG signals over six iterations, generating seven sub-bands.
- The L_p norm was computed for each sub-band to extract signal features.
- Extracted features were classified using K-nearest neighbor (KNN) algorithms with cross-validation.
Main Results:
- The developed CAD system achieved a highest accuracy of 99.21% using ten-fold cross-validation.
- Leave-one-subject-out cross-validation yielded an accuracy of 97.2%.
- Wavelet-based features effectively captured the non-stationary nature of EEG signals for SZ detection.
Conclusions:
- A single-channel EEG wavelet-based CAD model demonstrates high efficacy in diagnosing schizophrenia.
- The system offers a promising, accurate, and objective method to complement clinical assessment.
- This approach can enhance diagnostic confidence and efficiency for clinicians dealing with schizophrenia.
