An optimized deep hybrid learning for multi-channel EEG-based driver drowsiness detection

• Published in: Biomedical signal processing and control 2025-01, Vol.99, p.106881, Article 106881
• Main Authors: Latreche, Imene, Slatnia, Sihem, Kazar, Okba, Harous, Saad
• Format: Article
• Language: English
• Subjects: Deep hybrid learning; Deep learning; Driver drowsiness; Electroencephalograph (EEG); Hyperparameters optimization; Machine learning; Optuna; Preprocessing optimization
• ISSN: 1746-8094
• DOI: 10.1016/j.bspc.2024.106881

•Selecting the optimal set of preprocessing parameters that can enhance the classification results using the Random Search Optimization method.•Implementing multiple CNN architectures and selected the optimal one based on the mean accuracy of 10-fold cross-validation evaluation method.•Applying the automatic Hyperparameter tuning framework ‘Optuna’ to identify the optimal set of the CNN hyperparameters.•Fusing the CNN with Machine Learning classifiers (Deep Hybrid Learning).•Benefit from the power of the CNN in automatically extracting EEG features and the advantages of the ML classifiers. Driver drowsiness is a severe issue that has contributed to numerous fatal accidents and injuries. Thus, detecting driver drowsiness is an important task that has been the subject of intensive research in recent years. There have been numerous proposed physiological signals for detecting driver drowsiness. However, the Electroencephalographic (EEG) signal is the most commonly used due to its direct relationship with drowsiness and its simplicity of acquisition. Recently, different Machine Learning (ML) and Deep Learning (DL) models have been proposed to detect driver drowsiness. This study utilized a publicly accessible dataset containing twelve healthy participants. Reading numerous research papers, we determined no specific EEG-based drowsiness preprocessing parameter values. Consequently, as a first step, and for the first time, to our knowledge in this field, we applied an optimization algorithm to determine the optimal preprocessing parameter values using a CNN model and accuracy as the objective function. The obtained results demonstrated the importance of selecting the correct values, as the mean accuracy score increased from 91% before optimization to 95% after optimization for the proposed CNN. The training time has been reduced. Also, as a second step, we have used the Optuna Hyperparameter optimization framework to select the optimal CNN Hyperparameters, which increased the mean accuracy from 95% to 97%. Finally, to take advantage of Deep Hybrid learning, we have fused the CNN “features extractor” with seven ML classifiers, with the CNN-SVM classifier achieving the highest average accuracy of 99.9%, and the training time has been reduced to a shallow value.