Aging Detection Based on Dynamic State Transitions in Instantaneous Hilbert-Based Spatio-Temporal EEG Features

The spatial distribution of electroencephalography (EEG) oscillatory power and its temporal transitions are widely recognized as indicators of cognitive processes and pathological conditions, termed as microstates. These microstates reflect whole-brain neural network dynamics, including deep brain regions, and are closely associated with large-scale networks such as the default mode network. The conventional approach to microstate analysis relies on the envelope of EEG oscillations, which corresponds to the instantaneous amplitude. In this study, we aimed to extend conventional microstate analysis by integrating the instantaneous amplitude (power component) and instantaneous frequency data derived from the Hilbert transform. While our previous studies demonstrated that instantaneous frequency also reflects brain activity, this study highlights that integrating both features enables a more comprehensive assessment of aging effects. This integration allows the identification of brain states that cannot be detected using conventional power-based microstate analysis. Our findings suggest that this approach expands and enhances traditional microstate analysis and offers a novel index for detecting brain states. This method has the potential to provide new insights into neural network dynamics and can be applied to the study of cognitive processes and pathological conditions.