Tetromino pattern based accurate EEG emotion classification model.

Area of Science:

• Neuroscience
• Computer Science
• Signal Processing

Background:

• Emotion recognition from electroencephalogram (EEG) signals is a growing research area.
• Accurate classification of emotions using EEG data is crucial for various applications.
• Existing methods often require complex feature engineering or lack generalizability.

Purpose of the Study:

• To develop and validate a novel, automated emotion classification model for EEG signals.
• To introduce a game-based feature generation method inspired by Tetris, named Tetromino.
• To achieve high accuracy in emotion recognition using EEG data.

Main Methods:

• EEG signals were decomposed using discrete wavelet transform (DWT).
• Novel textural features were generated from DWT sub-bands using the Tetromino method.
• Maximum relevance minimum redundancy (mRMR) selected features, followed by support vector machine (SVM) classification and a voting-based ensemble method.

Main Results:

• The Tetromino-based model achieved 100% accuracy on the DREAMER and GAMEEMO datasets.
• Over 99% classification accuracy was obtained on the DEAP dataset.
• The model demonstrated superior performance compared to state-of-the-art techniques.

Conclusions:

• The Tetromino pattern-based EEG signal classification model shows significant success in emotion recognition.
• The proposed method offers a highly accurate and potentially clinically applicable approach to emotion detection.
• Further validation with diverse datasets is recommended for broader clinical application.