Related Experiment Video
Updated: Sep 26, 2025
09:35
Automatic Detection of Highly Organized Theta Oscillations in the Murine EEG
Published on: March 10, 2017
9.3K
CGP17Pat: Automated Schizophrenia Detection Based on a Cyclic Group of Prime Order Patterns Using EEG Signals
Emrah Aydemir1, Sengul Dogan2, Mehmet Baygin3
1Department of Management Information Systems, Management Faculty, Sakarya University, Sakarya 54050, Turkey.
Healthcare (Basel, Switzerland)
|April 23, 2022
Summary
A novel machine learning method effectively classifies schizophrenia using electroencephalogram (EEG) signals. The cyclic group of prime order pattern (CGP17Pat) achieved high accuracy, demonstrating its potential for schizophrenia detection.
Area of Science:
- Neuroscience
- Machine Learning
- Biomedical Engineering
Background:
- Machine learning models are increasingly utilized for diagnosing complex neurological conditions like schizophrenia.
- Accurate and early diagnosis of schizophrenia is crucial for effective patient management and treatment.
Purpose of the Study:
- To introduce an effective, hand-modeled classification method for schizophrenia detection.
- To develop an automated schizophrenia detection model using electroencephalogram (EEG) signals.
Main Methods:
- Utilized a public EEG dataset and developed a novel feature extractor named cyclic group of prime order pattern (CGP17Pat).
- Employed iterative neighborhood component analysis (INCA) for feature selection and k-nearest neighbors (kNN) for classification.
- Implemented 10-fold cross-validation and leave-one-subject-out (LOSO) validation, with iterative hard majority voting for final results.
Main Results:
- The CGP17Pat-based model achieved an accuracy of 99.91% with 10-fold cross-validation.
- The model attained 84.33% accuracy using the leave-one-subject-out (LOSO) validation strategy.
Conclusions:
- The developed cryptologic pattern (CGP17Pat) demonstrates high classification ability for schizophrenia detection using EEG data.
- The findings suggest the proposed method is a promising tool for automated schizophrenia diagnosis.
