Deep Sleep Classification via EEG Signal Criticality: A Passive BCI Approach for Sleep-Improvement Neurofeedback

2026-06-11

AI Summary

This study proposes a high-accuracy automatic staging method for deep sleep (N3 stage) based on criticality features derived from electroencephalography (EEG), aimed at enabling intention-free closed-loop neurofeedback interventions. For the first time, criticality indices extracted via detrended fluctuation analysis (DFA) are employed for deep sleep identification. The approach integrates UMAP manifold learning with multiple classifiers and is validated on 347,232 EEG segments from 290 older women. Results demonstrate that Naïve Bayes significantly outperforms both deep neural networks and random forests in this nonlinear manifold task, achieving a mean balanced accuracy of 87.17%. This work establishes a novel paradigm for state-dependent neurofeedback by leveraging EEG criticality as a robust biomarker for deep sleep detection.

Abstract

Automated sleep staging is a fundamental application of passive Brain-Computer Interfaces (pBCI), decoding spontaneous neural states to enable closed-loop interventions independent of user intent. This study evaluates criticality features derived from Detrended Fluctuation Analysis (DFA) for the specific identification of deep sleep (N3). We analyzed $347,232$ EEG epochs from $290$ older women using UMAP manifold learning to visualize state transitions. Subsequently, six classifiers were benchmarked via 10-fold cross-validation, using balanced accuracy to determine the optimal "state-sensing" engine for neurofeedback.Naive Bayes achieved the highest mean balanced accuracy ($87.17\% \pm 0.24\%$), significantly outperforming a fully connected deep neural network (FNN: $81.58\%$) and Random Forest ($80.97\%$). Linear models (LDA: $57.21\%$; SVM: $51.01\%$) performed poorly, indicating that DFA-derived criticality features reside on a distinct, non-linear manifold. Probabilistic decoding of EEG criticality provides a high-accuracy sensing mechanism for pBCIs. This robust classification pipeline supports the development of state-dependent neurofeedback, such as targeted auditory stimulation, to enhance cognitive recovery.

Problem

• deep sleep classification
• EEG criticality
• passive BCI
• sleep staging
• neurofeedback

Innovation

• EEG criticality
• Detrended Fluctuation Analysis
• passive BCI
• deep sleep classification
• UMAP manifold learning