Abstract

A major challenge in cognitive neuroscience is developing reliable diagnostic tools for Disorders of Consciousness (DoC). Detecting dynamic brain connectivity configurations holds great promise for advancing diagnostics. Evidence indicates that certain fMRI-derived connectivity patterns are closely tied to the level of consciousness. However, their clinical utility remains constrained by practical limitations. In this study, we introduce EEG-based brain states as a real-time, bedside tool for detecting periods of enhanced brain activity in DoC patients. We analyzed data from 237 patients with chronic and acute DoC from three different centers and identified five EEG functional connectivity recurrent brain patterns. The occurrence probabilities of these patterns were strongly correlated with patients' levels of consciousness. High-entropy patterns were found exclusively in healthy participants, while low-entropy patterns became more prevalent with increasing DoC severity, crucially predicting individual recovery outcomes. To assess the real-time applicability of this approach, we conducted tests demonstrating reliable, real-time estimation of patient brain patterns, confirming the feasibility of bedside detection. Our findings highlight the potential of EEG for real-time, bedside monitoring of brain dynamical connectivity patterns, significantly deepening our understanding of the neural dynamics underlying consciousness and paving the way for future discoveries in brain state research.