eeg-criticality-deep-sleep-classification-neurofeedback

name	eeg-criticality-deep-sleep-classification-neurofeedback
description	Deep Sleep Classification via EEG Signal Criticality using Detrended Fluctuation Analysis (DFA) for passive Brain-Computer Interface (pBCI) neurofeedback applications. Probabilistic decoding of EEG criticality features for state-dependent sleep improvement interventions.
version	1.0.0
category	neuroscience
authors	["Stanisław Narębski","Tomasz Komendziński","Tomasz M. Rutkowski"]
arxiv_id	2606.13017
published	2026-06-11T00:00:00.000Z
activation_keywords	["EEG criticality","deep sleep classification","DFA detrended fluctuation analysis","passive BCI","neurofeedback","sleep staging","N3 sleep","state-dependent intervention","probabilistic decoding","manifold learning"]
related_skills	["eeg-foundation-model-adapters","eeg-test-time-adaptation-benchmark","bci-rehabilitation-protocols","sleep-like-consolidation-llm"]

EEG Criticality Deep Sleep Classification for Neurofeedback

Overview

This methodology presents a probabilistic decoding approach for deep sleep (N3) identification using EEG signal criticality features derived from Detrended Fluctuation Analysis (DFA). The framework enables passive Brain-Computer Interface (pBCI) applications for state-dependent neurofeedback interventions such as targeted auditory stimulation for cognitive recovery enhancement.

Core Innovation

• Criticality-Based Features: DFA-derived scaling exponents capture sleep state transitions
• Manifold Visualization: UMAP reveals non-linear criticality manifold structure
• Bayesian Classification: Naive Bayes achieves 87.17% accuracy, outperforming deep networks
• Clinical Dataset: 347,232 EEG epochs from 290 older women (real-world validation)

Scientific Foundation

Criticality Theory in Sleep Dynamics

The brain exhibits critical dynamics during sleep, characterized by:

• Scale-free fluctuations: Long-range temporal correlations
• Phase transitions: Sharp changes between sleep stages
• Neural avalanches: Cascading activity patterns
• Optimal information processing: Balance at critical point

Detrended Fluctuation Analysis (DFA)

DFA quantifies self-similarity and scaling behavior:

1. Algorithm:

   Input: Time series x(t) of length N
   Integrate: y(k) = Σᵢ₌₁ᵏ (xᵢ - ⟨x⟩)
   
   For window sizes n:
   - Divide into N/n segments
   - Fit local trend yₙ(k) in each segment
   - Compute fluctuation: F(n) = √(1/N Σₖ (y(k) - yₙ(k))²)
   
   Scaling exponent α: F(n) ~ n^α
   

2. Interpretation:

   • α < 0.5: Anti-correlated (subcritical)
   • α = 0.5: Random (uncorrelated)
   • 0.5 < α < 1.0: Long-range correlated (critical)
   • α = 1.0: 1/f noise (pink noise)
   • α > 1.0: Non-stationary (supercritical)

3. Sleep Stage Signatures:

   • Wake: α ≈ 0.5-0.7 (mild correlations)
   • N1 (light sleep): α ≈ 0.6-0.8
   • N2 (intermediate): α ≈ 0.7-0.9
   • N3 (deep sleep): α ≈ 0.9-1.2 (strong correlations)