Deep Hybrid CNN-BiLSTM-Attention Model for EEG Classification Using Wavelet Features.

A deep hybrid model combining convolutional neural networks, bidirectional long short-term memory networks, and attention mechanisms achieved 99.46% accuracy in classifying EEG recordings from experienced meditation practitioners into resting or mantra-listening brain states. Wavelet-based time-frequency features were extracted from EEG data recorded before and during auditory mantra stimulation. The model outperformed simpler architectures including CNN alone (76.92%), LSTM alone (75.30%), CNN+LSTM (84.62%), and CNN+BiLSTM (88.65%). Receiver operating characteristic analysis confirmed high discriminative capability with an AUC near 1.0. The approach demonstrates that combining convolutional, recurrent, and attention methods substantially improves spatial-temporal feature learning for distinguishing brain states.

Study at a glance

Abstract

Electroencephalography (EEG) is a popular non-invasive method for studying brain dynamics because of its excellent temporal resolution. However, the non-stationarity, intersubject variability and class imbalance of EEG data, make it difficult to automatically discriminate between brain states that correspond to various cognitive or sensory circumstances. With the use of a deep hybrid convolutional neural networks (CNNs) and bidirectional long short-term memory (BiLSTM) networks with an attention architecture intended to improve discriminative learning from wavelet-based time-frequency features, this study attempts to categorise EEG recordings into discrete brain states recorded prior to and during auditory (mantra) stimulation. Experienced practitioners (mean age: 37 ± 6 years; mean practice: 5 years) had their EEG data recorded in two different experimental settings: (a) When they were at rest before the auditory stimulus and (b) while they were listening to mantras. Each segment's time-frequency representations were produced using wavelet transforms and fed into a hybrid model that combined convolutional, recurrent and attention layers. To guarantee steady convergence, adaptive learning rate scheduling and early stopping were used in the model optimisation process. With CNN (76.92%), long short-term memory (LSTM) (75.30%), CNN+LSTM (84.62%) and CNN+BiLSTM (88.65%), baseline models performed moderately. The suggested CNN-BiLSTM-Attention model achieved an independent test accuracy of 99.46%, greatly outperforming all baselines. High discriminative capability was confirmed by the receiver operating characteristic (ROC) analysis, which produced an AUC near 1.0. The inclusion of convolutional, recurrent and attention methods greatly improves spatial-temporal feature learning, as demonstrated by the suggested framework's ability to distinguish between resting and during mantra EEG states. These results demonstrate the model's resilience and possible use in neurophysiological monitoring and real-time cognitive state detection.