Related Experiment Video
Updated: Nov 16, 2025
09:57
Author Spotlight: Advancing Pediatric Epilepsy Surgery in Children Through Novel Biomarkers and Enhanced Localization
Published on: September 20, 2024
3.1K
A hybrid deep neural network for classification of schizophrenia using EEG Data
Jie Sun1, Rui Cao2, Mengni Zhou3
1College of Information and Computer, Taiyuan University of Technology, Taiyuan, China.
Scientific Reports
|February 26, 2021
Summary
This study introduces a novel deep learning method for schizophrenia detection using electroencephalography (EEG) signals. Fuzzy entropy features combined with a hybrid deep neural network achieved 99.22% accuracy, significantly improving diagnosis.
Area of Science:
- Neuroscience
- Artificial Intelligence
- Medical Diagnostics
Background:
- Schizophrenia poses a significant global health burden, necessitating accurate and timely diagnostic tools.
- Electroencephalography (EEG) signals offer a non-invasive window into brain activity, but extracting relevant features for complex conditions like schizophrenia remains challenging.
Purpose of the Study:
- To identify optimal features from EEG signals for improved classification of schizophrenia.
- To develop and validate a hybrid deep learning model for enhanced diagnostic accuracy in schizophrenia patients.
Main Methods:
- EEG time-series data were preprocessed, and time- and frequency-domain features were extracted.
- Extracted features were transformed into RGB images to incorporate spatial information.
- A hybrid deep neural network (DNN) combining Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) was employed for classification.
Main Results:
- Fuzzy entropy (FuzzyEn) demonstrated superior significance over Fast Fourier Transform (FFT) for brain topography analysis.
- The proposed deep learning model achieved an average accuracy of 99.22% using FuzzyEn features.
- The model achieved an average accuracy of 96.34% using FFT features, outperforming existing state-of-the-art methods.
Conclusions:
- Fuzzy entropy is a highly effective feature for EEG-based schizophrenia detection.
- The hybrid DNN model integrating FuzzyEn features offers a significant advancement in classifying schizophrenia patients.
- This approach represents a substantial improvement over current methods for EEG signal analysis in schizophrenia.
