Skip to content

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

Tony Bayan, Daisy Das, Nabamita Deb

Annals of Neurosciences January 23, 2026 DOI: 10.1177/09727531251396337 (opens in new tab)

Read Full Text
Relevant Not relevant

Study at a glance

AI-extracted from the abstract
Characteristics Observational study Peer reviewed
Population Experienced meditation practitioners (mean age 37 ± 6 years; mean practice 5 years)
Topics Meditation
Keywords Om Mantra Wavelet
Key findings The proposed CNN-BiLSTM-Attention model achieved 99.46% accuracy in classifying EEG recordings into resting and mantra-listening brain states, greatly outperforming baseline models.

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.

Comparable studies

Other observational and cohort studies on meditation, most cited first.

Study Year Design Participants
Using Self-Report Assessment Methods to Explore Facets of Mindfulness Undergraduate students 2006 Observational cohort
Effectiveness of a meditation-based stress reduction program in the treatment of anxiety disorders Patients with generalized anxiety disorder or panic disorder with or without agoraphobia 1992 Observational cohort n = 22
Meditation experience is associated with increased cortical thickness Participants with extensive Insight meditation experience and matched controls 2005 Observational cohort n = 20
Meditation experience is associated with differences in default mode network activity and connectivity Experienced meditators and matched meditation-naive controls 2011 Observational cohort
Long-term meditators self-induce high-amplitude gamma synchrony during mental practice Long-term Buddhist practitioners and controls 2004 Observational cohort

Citations in the library

Cites 5

Related reading

Explore topics

Meditation